Jump to main content.

Current Good Manufacturing Practice in Manufacturing, Packing, or Holding Dietary Ingredients and Dietary Supplements

 [Federal Register: March 13, 2003 (Volume 68, Number 49)]
[Proposed Rules]
[Page 12207-12256]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr13mr03-32]
 
[[pp. 12207-12256]]
Current Good Manufacturing Practice in Manufacturing, Packing, or 
Holding Dietary Ingredients and Dietary Supplements

[[Continued from page 12206]]

[[Page 12207]]

problems are associated with maintenance, cleaning, or sanitizing 
operations. Batch and lot information, as we stated earlier, will let 
you identify batches or lots that may have been affected by any 
equipment or utensil that was improperly maintained, cleaned, or 
sanitized.
    Proposed Sec.  111.50(d) and (e) would set forth your quality 
control unit's responsibilities regarding batch production records. 
These responsibilities relate to not only the review but the 
documentation of their review and decisions about whether a batch could 
be reprocessed. As we noted in our discussion of proposed Sec.  111.37, 
the quality control unit has special knowledge and expertise to 
determine if a batch is produced correctly, that those records are 
complete, and that it is appropriate to reprocess a batch. The quality 
control unit also serves as a quality control check that the batch 
production record accurately follows the master manufacturing record. A 
quality control unit review of batch production records could have 
detected and corrected the previously discussed manufacturing error 
caused by use of the dietary ingredient with the incorrect selenium. 
Therefore, the review and documentation by the quality control unit of 
batch production records provides the necessary quality assurance to 
prevent the production of an adulterated dietary ingredient or dietary 
supplement.
    Specifically, proposed Sec.  111.50(d) would require your quality 
control unit to review in accordance with Sec.  111.37(b)(5) the batch 
production record. If a batch production record deviates from the 
master manufacturing record, including any deviation from 
specifications, proposed Sec.  111.50(d)(1) would require your quality 
control unit to conduct a material review and make a disposition 
decision and record any decision in the batch production record. 
Proposed Sec.  111.50(d)(2) would instruct your quality control unit to 
not approve and release for distribution any batch of dietary 
ingredient or dietary supplement that does not meet all specifications.
    Proposed Sec.  111.50(e) would require your quality control unit to 
document in accordance with Sec.  111.37(c) the review performed in 
accordance with proposed Sec.  111.50(d). The proposal would require 
the quality control unit to document this review at the time it does 
the review and would require the review and documentation to include, 
but would not limit them to, the following:
    ? Review of component, dietary ingredient, and dietary 
supplement receiving records including review of testing and 
examination results;
    ? Identification of any deviation from the master 
manufacturing record that may have caused a batch or any of its 
components to fail to meet specifications identified in the master 
manufacturing record;
    ? Records of investigations, conclusions, and corrective 
actions performed in accordance with proposed Sec.  111.50(d); and
    ? The identity of the person qualified by training and 
experience who performed the investigation in accordance with proposed 
Sec.  111.50(d).
    Proposed Sec.  111.50(f) would prohibit you from reprocessing a 
batch that deviates from the master manufacturing record unless your 
quality control unit approves it for reprocessing. Proposed Sec.  
111.50(f) also would prohibit you from reprocessing a dietary 
ingredient or dietary supplement if it is rejected because of 
contamination with microorganisms of public health significance or 
other contaminants, such as heavy metals because you cannot rely on 
reprocessing to correct public health concerns that a product with 
pathogens and/or heavy metals would present.
    Proposed Sec.  111.50(g) would require that you meet all 
specifications established in the master manufacturing record for any 
batch of dietary ingredient or dietary supplement that is reprocessed 
and would require your quality control unit to evaluate and approve the 
batch before releasing for distribution. This requirement is intended 
to ensure that a reprocessed batch is not subject to any lesser 
specifications than are otherwise applicable to a nonreprocessed batch. 
Proposed Sec.  111.50(g) also would require that you document the 
results of the quality control unit's reevaluation in the batch 
production record.
    Proposed Sec.  111.50(h) would require that you collect 
representative reserve samples of each batch of dietary ingredient or 
dietary supplement and to keep the reserve samples for 3 years from the 
date of manufacture for use in appropriate investigations including, 
but not limited to, consumer complaint investigations to determine 
whether, for example, the dietary ingredient or dietary supplement 
associated with a consumer complaint failed to meet any of its 
specifications for identity, purity, quality, strength, and 
composition. Reserve samples also may prove helpful in investigating 
possible tampering or counterfeiting of your products. We invite 
comment on whether we should require, in a final rule, that you 
identify each reserve sample with the batch number so that you can 
readily identify the correct reserve sample in the event that there is 
a problem with a particular batch.
    Proposed Sec.  111.50(i) would require that you keep your batch 
production records in accordance with proposed Sec.  111.125. The batch 
production records in addition to the master manufacturing records will 
ensure that a complete history of each batch of dietary ingredient or 
dietary supplement is available for your review in the event that a 
problem arises with a particular batch. These records also are 
necessary to demonstrate compliance with the CGMP and quality control 
procedures.
6. What Requirements Apply to Laboratory Operations? (Proposed Sec.  
111.60)
    Proposed Sec.  111.60 would establish various requirements for 
laboratory operations. Proposed Sec.  111.60(a) would require that you 
use adequate laboratory facilities to perform any necessary tests or 
examinations to determine that components, dietary ingredients, and 
dietary supplements you receive meet specifications; that 
specifications are met during inprocess as specified in the master 
manufacturing record; and that the dietary ingredients and dietary 
supplements you manufacture meet their specifications.
    One comment to the ANPRM recommended that the regulations related 
to laboratory operations apply to laboratory facilities located and 
operated within a company and those facilities that a company may 
contract with that are located elsewhere. Proposed Sec.  111.60(a) 
would apply to laboratory facilities generally and is not restricted to 
laboratory facilities located and operated within a company. In other 
words, even if you hire a private laboratory to perform various tests 
for you, proposed Sec.  111.60(a) would require that you make sure that 
the private laboratory's facilities are adequate to perform whatever 
tests are necessary. The most important point in proposed Sec.  
111.60(a), however, is not where the facility is located, but whether 
the laboratory facility is adequate for the tests and examinations that 
need to be done.
    Proposed Sec.  111.60(b)(1) would require that you establish and 
follow laboratory control processes that the quality control unit has 
approved. For example, under proposed Sec.  111.60(b)(1)(i) and 
(b)(1)(ii), the laboratory control processes would include use of 
criteria for selecting appropriate testing and examination methods and 
for establishing appropriate specifications. Specifications play an 
important role in CGMP's because they may help

[[Page 12208]]

determine whether a dietary ingredient or dietary supplement is 
adulterated.
    Criteria for establishing appropriate specifications must be 
specific to the component, dietary ingredient, or dietary supplement. 
The specifications are the parameters that you must meet. For example, 
for ascorbic acid, your specifications would include all the criteria 
that you want your incoming dietary ingredient or for your finished 
product to meet. For example, you might establish criteria for the 
appearance, color, odor, identity using one or more tests, heavy metals 
(e.g., lead, arsenic, mercury), and organic volatile impurities.
    Similarly, criteria for selecting appropriate test and examination 
methods include parameters such as type of tests and examinations 
needed based on the component you receive. For example, you might use 
morphological characters and organoleptic characteristics in some cases 
to identify botanical dietary ingredients at the time of collection or 
for unprocessed botanicals. When sufficient morphological characters 
are present to separate the plant species from other plant species, an 
accurate identification can be made since morphological characters are 
the sole basis of distinguishing most of the world's plant species. 
However, unprocessed botanicals that do not contain all the plant parts 
necessary to include adequate morphological characters to assure the 
correct species should have other identity aids or tests to assure the 
identity of the botanical. It is possible to use only a picture as an 
identity standard for whole fresh Ginkgo leaf from a cultivated field 
because the Ginkgo leaf is not easily confused with the leaf shape, 
venation, and color of other leaves that could be present in the field. 
In contrast, powdered Ginkgo leaf is a different form of the dietary 
ingredient and would require microscopic and/or chemical analysis. 
Ginkgo extracts have no morphological or anatomical features, and it is 
possible that extracts may include a number of chemical compounds at 
different ratios and concentrations that would require a different 
chemical test to assure the identity of the dietary ingredient. 
Botanical dietary ingredients that come from wild rather than 
cultivated sources may grow among and be unintentionally harvested with 
``poisonous'' plants; therefore, an identity test also would need to 
show whether a botanical dietary ingredient is adulterated with another 
substance or a poisonous plant.
    To illustrate this point, a specification may contain a simple 
identity test, and these tests may show whether a dietary ingredient is 
adulterated with another substance or is a poisonous plant that should 
not be ingested. Misidentification or a mixup of botanical ingredients 
can cause a product to be adulterated (Refs. 6 and 69 through 73). 
Heavy metals may contaminate botanical and natural-occurring 
ingredients if a plant is grown and harvested in an area contaminated 
with heavy metals or even processed in a contaminated area (Refs. 74 
and 75). Pesticides also may contaminate botanical ingredients; this 
occurs in rural areas where the botanical plant grow alongside 
commercial crops (Ref. 64). Therefore, you must consider what criteria 
you need to include for the types of testing that are needed, for 
example, for heavy metal or pesticide contamination, or identity 
testing criteria for selecting appropriate test methods, for example, 
whether to use organoleptic or chemical analyses for identity testing. 
In addition, you must establish criteria for specifications for the 
tests and examinations used. Establishing such criteria for 
specifications and appropriate test and examination methods will 
provide you with internal processes that will help prevent 
misidentification and contamination.
    Proposed Sec.  111.60(b)(1)(iii) would require your laboratory 
control processes to include use of sampling plans for obtaining 
representative samples of:
    ? Components, dietary ingredients, and dietary supplements 
received;
    ? Inprocess materials during the batch manufacturing when 
testing or examination is required in the master manufacturing record;
    ? Each batch of dietary ingredient or dietary supplement 
manufactured to determine that the dietary ingredient or dietary 
supplement meets specifications;
    ? Packaging and labels received to determine that the 
materials meet specifications; and
    ? Each batch of packaged and labeled dietary ingredients or 
dietary supplements to ensure that the label specified in the master 
manufacturing record has been applied.
    For example, a representative sample is important to being able to 
have an adequate sample to detect contamination. Contamination may not 
be distributed evenly throughout a product and may not be detected 
without a representative number of units. Determining the size of a 
representative sample is important because the sample size must be 
large enough to meet your testing needs for specific types of 
components, dietary ingredients, or dietary supplements, and packaging 
and labels. Your sampling plans should include reserve samples, too, 
because reserve samples will enable you to investigate and identify 
possible manufacturing problems in the future. The proposal would not 
specify any particular sampling plan; it would leave such details to 
your discretion so that you can develop a sampling plan that suits your 
products and your testing needs.
    Proposed Sec.  111.60(b)(iv) through (b)(vi) would require the 
laboratory control processes to include:
    ? Use of criteria for selecting standard reference materials 
used in performing tests and examinations. An authenticated plant 
reference material may be used as standard reference material in 
performing certain organoleptic examinations. An authenticated plant 
reference material is material that has been authenticated as the 
correct plant species and correct plant part(s) by a qualified plant 
taxonomist. As described earlier in this document, an organoleptic 
examination may be an appropriate examination to confirm plant identity 
when sufficient morphological characters are present to separate the 
plant species from other plant species. For microscopic and chemical 
tests, a reference material is a highly purified compound that is well 
characterized, and you would use the reference material to perform 
tests including calibration tests. In general, there are two types of 
reference materials: (1) Compendial reference standards that do not 
require characterization; and (2) noncompendial standards. 
Noncompendial standards should be of the highest purity that can be 
obtained by reasonable effort and should be thoroughly characterized to 
assure their identity, purity, quality, and strength. Ideally, you 
should use compendial reference standards whenever possible, but if no 
compendial reference standard exists, you should establish 
appropriately characterized inhouse materials prepared from 
representative lots;
    ? Use of appropriate test method validations. Test method 
validation determines whether a newly-developed or existing test method 
is accurate, precise, and specific for its intended purpose. We have 
discussed previously the terms ``accurate'' and ``precise.'' Validation 
involves evaluating the test method on multiple occasions or in 
multiple test facilities. Official methods, such as AOAC International 
methods, are validated in collaborative studies using several 
laboratories under identical conditions. The AOAC International methods 
that are validated

[[Page 12209]]

in collaborative studies often are often cited as ``official validated 
methods.'' If you modify an officially validated method, you should 
document the reason for the modification and have data to show that the 
modified method produced results that are at least as accurate and 
reliable as the established method for material being tested. Further, 
you should have complete records of any testing and standardization of 
laboratory reference standards, reagents, and standard solutions that 
you use in your laboratory operations. Proposed Sec.  111.25(b)(1) 
would require calibration of laboratory instruments, apparatus, gauges, 
and recording devices. Validated methods also exist in official 
compendia for vitamins, minerals, and several botanicals, so you should 
use validated methods whenever available. You may use validated methods 
that can be found in official references, such as AOAC International, 
USP and others. Other method validations are conducted using two or 
three laboratories or in a single laboratory by repeating the same test 
multiple times. Official and nonofficial method validations use similar 
performance parameters in conducting method validations. If an official 
validated method does not exist in an official reference, the method 
you use may be validated by using multiple tests at your laboratory or 
multiple laboratories performing the same test to document that the 
intended use of the method is consistently fulfilled. You must validate 
that the official or nonofficial method works under your conditions of 
use in your setting. You also should conduct day-to-day validations of 
the method that you use, whether it is an official validated method or 
a less-formal validated method, under the conditions of use to ensure 
that the method will provide the information you need to ensure that 
your dietary ingredient or dietary supplement has the identity, purity, 
quality, strength, and composition that it is supposed to have and is 
thus not adulterated. Consistent, day-to-day test recoveries for the 
reference material are one indicator that the analytical method is 
working. There are at least two references that describe test method 
validation performance parameters: (1) Performance parameters for 
chromatographic methods are described in ``Reviewer Guidance, 
Validation of Chromatographic Methods'' (Center for Drug Evaluation and 
Research, FDA, November 1994) (Ref. 76); and (2) International 
Conference on Harmonisation (ICH); Draft Guidance on Specifications: 
Test Procedures and Acceptance Criteria for Biotechnological/Biological 
Products (63 FR 31506, June 9, 1998); and
    ? Use of test methods in accordance with established 
criteria. Your process for performing test methods criteria must 
include sufficient detail, including the material you are testing, the 
purpose of the test, and the test method. The description of the test 
method criteria must include any reagents used and preparation 
instructions, apparatus required, any instructions for preparing the 
sample to be tested, and instructions for conducting the examination. 
For example, if you receive components of plant origin from an outside 
source, your specifications must indicate that you test those 
components to verify that they are not contaminated with adulterants of 
vegetable origin and to determine that the microscopic examination 
method is appropriate for use. Further, you may decide that the AOAC 
International Official Method 961.01 entitled ``Adulterants in Spices'' 
is the appropriate analytical method to detect the contaminant which is 
a method to detect adulterants of vegetable origin in spices. Your test 
methods criteria must specify the component, dietary ingredient, or 
dietary supplement to be tested, and what specifically to test for, 
e.g., the identity of the component, dietary ingredient, or dietary 
supplement, or a specific contaminant. The method criteria must provide 
detailed information about performing the analysis (i.e., the reagent 
solutions needed and their preparation, the type of microscope and 
other equipment required, preparing the sample, and examination 
instructions). The proposed rule would not require that you test for 
any specific substance and would not require a specific test for a 
substance, so you would be able to evaluate what the most appropriate 
test would be for the component, dietary ingredient, or dietary 
supplement and to use the test methods that are suited to your products 
and your manufacturing needs. Your test methodology must be specific 
for the component, dietary ingredient, or dietary supplement and the 
specifications you have established.
    Proposed Sec.  111.60(b)(2) and (b)(3) would apply to documentation 
and recordkeeping for your laboratory operations. Proposed Sec.  
111.60(b)(2) would require the person who conducts the testing and 
examination to document, at the time of performance, that they followed 
the laboratory method and the testing and examination results. Proposed 
Sec.  111.60(b)(3) would require that you keep laboratory testing and 
examination records in accordance with proposed Sec.  111.125. 
Laboratory records are necessary to ensure compliance with established 
specifications and to demonstrate compliance with the CGMP and quality 
control processes.
    Proposed Sec.  111.60(c) would require that you verify that the 
laboratory testing methodologies are appropriate for their intended 
use.
    Proposed Sec.  111.60(d) would require that you identify and use 
the appropriate validated testing method to use for each established 
specification for which testing is required to determine whether the 
specification is met. In other words, the proposal would recognize that 
requiring that you have testing methods is not sufficient alone; you 
must also use those testing methods to prevent the adulteration of 
dietary ingredients or dietary supplements.
    We invite comment on whether we should require, in a final rule, 
written procedures for your laboratory operations and should require 
that the person who performs the laboratory processes document, at the 
time of performance, that the laboratory processes were performed. If 
comments assert that written procedures are necessary, comments should 
include an explanation of why the requirement is necessary to prevent 
adulteration including how such a requirement would ensure the 
identity, purity, quality, strength, and composition of the dietary 
ingredient or dietary supplement. Conversely, if comments assert that 
written procedures are not necessary, comments should include an 
explanation of why the requirement is not necessary including how, in 
the absence of the requirement, one can prevent adulteration and ensure 
the identity, purity, quality, strength, and composition of the dietary 
ingredient or dietary supplement. Further, we seek comment on whether 
any of the proposed requirements in this section are not necessary to 
prevent adulteration and to ensure the identity, purity, quality, 
strength, and composition of the dietary ingredient or dietary 
supplement. If comments assert that certain provisions are not 
necessary, comments should include an explanation of why the 
requirement is not necessary including how, in the absence of the 
requirement, one can prevent adulteration and ensure the identity, 
purity, quality, strength, and composition of the dietary ingredient or 
dietary supplement. If comments agree that the proposed requirements 
are necessary for reasons other than those we have provided, the 
comments should so state and provide an explanation.

[[Page 12210]]

7. What Requirements Apply to Manufacturing Operations? (Proposed Sec.  
111.65)
    Proposed Sec.  111.65 would require that you take all necessary 
precautions to ensure that, during the manufacturing operations, you do 
not create a source of possible contamination and that specifications 
are consistently achieved.
    Under proposed Sec.  111.65(a), you must design or select equipment 
and processes to ensure that dietary ingredient or dietary supplement 
specifications are consistently achieved. Frequently, a computer or 
system of computers may control many or all stages of manufacturing 
operations such as mixing, producing tablets, and packaging. It is 
important that such systems and equipment function as expected to 
ensure that the dietary ingredient or dietary supplement contains a 
homogenous mixture, a tablet that is neither too hard or too friable, 
and that the packaging contains the correct dietary ingredient or 
dietary supplement. Equipment used in dietary ingredient or dietary 
supplement manufacture, packaging, and label operations must be, for 
example, of an appropriate size and installed properly in order to 
produce an unadulterated product. If not designed or installed 
properly, the equipment can result in a variety of problems. For 
example, a mixer for the blending of powdered ingredients will not 
properly perform its function if the blade is too small relative to the 
size of the mixer or not properly placed inside of the mixer. Such a 
mixer may produce an adulterated product because the dietary 
supplement, for example, is not of uniform composition and therefore 
would not be able to meet the specifications for purity, quality, 
strength, or composition in the final product. Thus, equipment design 
and selection is critical to ensure that you manufacture an 
unadulterated dietary ingredient or dietary supplement.
    Proposed Sec.  111.65(b) would require that you conduct all 
manufacturing operations in accordance with adequate sanitation 
principles. We discussed the importance of having adequate sanitation 
earlier and related it to the use of sanitary practices for employees, 
physical plant, and equipment.
    Proposed Sec.  111.65(c)(1) through (c)(11) would require that you 
take all the necessary precautions during the manufacture of dietary 
ingredients and dietary supplements to prevent contamination of 
components, dietary ingredients, and dietary supplements.
    Proposed Sec.  111.65(c)(1) would require that you perform 
manufacturing operations under conditions and controls that protect 
against the potential for microorganism growth and the potential for 
contamination. This would require that you conduct all operations in 
receiving, inspecting, transporting, segregating, preparing, 
manufacturing, packaging, sorting, and packing dietary ingredients and 
dietary supplements in accordance with appropriate and established 
sanitation procedures. Operations with poor sanitation in the 
production and processing environment can significantly increase the 
risk of contaminating components, dietary ingredients, or dietary 
supplements. Pathogenic microorganisms may be found on the floors and 
in the drains of the processing area and on all contact surfaces. 
Without good sanitary practices, any surface that comes in contact with 
components, dietary ingredients, and dietary supplements could be a 
potential source of microbial contamination. Thus, using appropriate 
sanitation procedures would provide conditions and controls to protect 
against potential contamination and microbial growth.
    Proposed Sec.  111.65(c)(2) would require that you wash or clean 
components that contain soil or other contaminants. This is a basic 
sanitation procedure to protect against contamination and microbial 
growth. Raw agricultural materials and other components that contain 
soil or other contaminants must be washed or cleaned as necessary. 
Water quality used for washing, rinsing, or conveying raw agricultural 
materials must be adequate for its intended use, both at the start and 
at the end of the processing operation, and should not contribute to 
the contamination of such materials.
    Proposed Sec.  111.65(c)(3) would require that you use water that 
meets the EPA's NPDW regulations or, where necessary, higher sanitary 
quality and that complies with all applicable Federal, State, and local 
regulations for water that is used in the manufacturing operation. If 
you reuse water that was used to remove soil or contaminants from 
components, the proposal would require that the reused water be safe 
and of adequate sanitary quality so that it does not become a source of 
contamination. Some manufacturing operations may require water of a 
higher sanitary quality than water that meets the NPDW regulations. For 
example, the fluoride or chloride levels in water meeting the NPDW 
regulations may interfere with certain capsule or tablet operations and 
a higher quality water such as distilled water may be necessary. This 
proposed requirement allows the manufacturer discretion in determining 
whether NPDW regulations or higher sanitary quality water is necessary 
for a manufacturing operation.
    Proposed Sec.  111.65(c)(4) would require that you perform 
chemical, microbiological, or other testing, as necessary, to prevent 
the use of contaminated components, dietary ingredients, and dietary 
supplements. You should consider identifying those areas in the 
processing and production areas where chemical, microbial, or other 
forms of contamination are most likely to occur. Chemical, microbial, 
or other testing is necessary to identify areas where sanitation 
measures have not been adequate or where products may become 
adulterated.
    Proposed Sec.  111.65(c)(5) would require that you sterilize, 
pasteurize, freeze, refrigerate, control hydrogen-ion concentration 
(pH), control humidity, control water activity, or use any other 
effective means to remove, destroy, or prevent the growth of 
microorganisms and to prevent decomposition. The measures you decide to 
use to remove, destroy or prevent the growth of microorganisms on or in 
your components, dietary ingredients, or dietary supplements must be 
appropriate under the conditions of manufacture, handling, and 
distribution. Such measures are necessary to prevent their adulteration 
and misbranding. Microorganisms include pathogenic bacteria that, if 
present would adulterate the product. In addition, decomposition may 
result in a change in the component, dietary ingredient, or dietary 
supplement strength; the consequence of not using the appropriate 
measure may be that the dietary ingredient or dietary supplement no 
longer meets specifications, and thus, would be adulterated under 
section 402(g) of the act and misbranded under section 403 of the act. 
By including the phrase, ``any other effective means,'' we provide you 
with discretion to decide which measures to use to destroy or prevent 
the growth of microorganisms and to prevent decomposition.
    Proposed Sec.  111.65(c)(6) would require that you hold components, 
dietary ingredients, and dietary supplements that can support the rapid 
growth of microorganisms of public health significance in a manner that 
prevents them from becoming adulterated.
    Proposed Sec.  111.65(c)(7) would require that you identify and 
hold any components, dietary ingredients, and dietary supplements, that 
require a material review and disposition decision, in a manner that 
protects the components, dietary ingredients, and dietary supplements 
against

[[Page 12211]]

contamination and mixups. A dietary ingredient or dietary supplement 
under this proposed rule would require a material review and 
disposition decision when the components, dietary ingredients, or 
dietary supplements deviate from specifications. As previously 
explained, the specifications established as production and process 
controls under proposed subpart E of part 111, are regulatory 
specifications. Thus, a deviation from such a specification means that 
the components, dietary ingredients, or dietary supplements may be 
adulterated. Any component, dietary ingredient, or dietary supplement 
that may be adulterated must be segregated from such material that 
meets specifications so that it does not become a source of 
contamination. The proposal would require that you hold these 
components, dietary ingredients, and dietary supplements in a manner 
that protects against contamination and mixups.
    Proposed Sec.  111.65(c)(8) would require that you perform 
mechanical manufacturing steps (such as cutting, sorting, inspecting, 
shredding, drying, grinding, blending, and sifting) by any effective 
means to protect the dietary ingredients or dietary supplements against 
contamination. Such steps must include consideration of cleaning and 
sanitizing contact surfaces, using temperature controls, and using time 
controls. For example, when blending components, if you use a mixer 
that has not been cleaned and sanitized, your blended material may 
become contaminated with microorganisms, including microbial pathogens. 
Thus, it is important to clean and sanitize your mixer before use.
    Proposed Sec.  111.65(c)(9) would require that you use effective 
measures, such as filters, traps, magnets, or electronic metal 
detectors, to protect against the inclusion of metal or other foreign 
material in your components, dietary ingredients, or dietary 
supplements. This proposed requirement is intended to exclude foreign 
and extraneous matter that would contaminate components, dietary 
ingredients, or dietary supplements. The purpose of this proposed 
requirement is not to exclude dietary ingredients that are intended to 
be used and that are of mineral origin.
    One comment to the ANPRM suggested that we require the use of 
effective measures to protect against the inclusion of metal or other 
extraneous material in dietary products when there is reason to suspect 
that the product is contaminated by metal or other extraneous material. 
The comment stated that manufacturers typically are able to identify 
the particular piece of equipment that is the source of the metal 
contamination.
    We disagree with the comment. The purpose behind proposed Sec.  
111.65(c)(9) is to ensure that no metal or foreign material becomes a 
source of possible contamination and not to establish mechanisms to be 
used after contamination has or is suspected to have occurred. We 
believe that the most practical way to protect against the inclusion of 
metal and foreign material is to require that you use effective 
measures during the manufacturing operations. The source of metal 
contamination is not limited to equipment and we previously emphasize 
the need to maintain equipment to prevent such contamination. Metal 
contamination also may occur during harvesting of natural products and 
use of utensils such as metal brushes. Therefore, because we believe 
that it is not possible to identify and eliminate all possible sources 
of metal contamination or to determine when measures would be necessary 
to eliminate such contamination, proposed Sec.  111.65(c)(9) would 
require that you use effective measures to protect against the 
inclusion of metal and foreign material for all your manufacturing 
operations.
    Proposed Sec.  111.65(c)(10) would require that you segregate and 
identify all containers for a specific batch of dietary ingredients or 
dietary supplements to identify their contents and, where necessary, 
the phase of manufacturing. This proposed requirement is intended to 
protect ingredients or dietary supplements from potential contamination 
or misuse during manufacturing or storage. Identifications of these 
items will enable you to determine accurately the status of all batches 
of dietary ingredients or dietary supplements during all stages of the 
manufacturing process, will help to prevent mixups in the addition of 
components or dietary ingredients to the dietary supplement and will 
facilitate prompt action if any problems in processing are identified.
    Proposed Sec.  111.65(c)(11) would require that you identify all 
processing lines and major equipment used during manufacturing and to 
indicate their contents, including the name of the dietary ingredient 
or dietary supplement and the specific batch or lot number and, when 
necessary, the phase of manufacturing. The same reasons given for 
proposed Sec.  111.65(c)(10) apply to this proposed requirement.
    Proposed Sec.  111.65(d) would require that you conduct a material 
review and make a disposition decision in accordance with proposed 
Sec.  111.35(i) for any component, dietary ingredient, or dietary 
supplement that fails to meet specifications or that is, or may be, 
adulterated. If the material review and disposition decision allows you 
to reprocess the component, dietary ingredient, or dietary supplement, 
proposed Sec.  111.65(d) would require that you retest or reexamine it 
to ensure that it meets specifications and is approved by the quality 
control unit.
    The person who performs the material review and disposition review 
required in accordance with this section would be required to document 
at the time of performance the results of the material review and 
disposition decision. In accordance with Sec.  111.50(d), such 
documentation must be maintained with the batch production record.
    We invite comment on whether we should require, in a final rule, 
that you establish and follow written procedures to implement the 
manufacturing operations required in proposed Sec.  111.65. If comments 
assert that written procedures are necessary, comments should include 
an explanation of why the requirement is necessary to prevent 
adulteration including how such a requirement would ensure the 
identity, purity, quality, strength, and composition of the dietary 
ingredient or dietary supplement. Conversely, if comments assert that 
written procedures are not necessary, comments should include an 
explanation of why the requirement is not necessary including how, in 
the absence of the requirement, one can prevent adulteration and ensure 
the identity, purity, quality, strength, and composition of the dietary 
ingredient or dietary supplement. Further, we seek comment on whether 
any of the proposed requirements in this section are not necessary to 
prevent adulteration and to ensure the identity, purity, quality, 
strength, and composition of the dietary ingredient or dietary 
supplement. If comments assert that certain provisions are not 
necessary, comments should include an explanation of why the 
requirement is not necessary including how, in the absence of the 
requirement, one can prevent adulteration and ensure the identity, 
purity, quality, strength, and composition of the dietary ingredient or 
dietary supplement. If comments agree that the proposed requirements 
are necessary for reasons other than those we have provided, the 
comments should so state and provide an explanation.

[[Page 12212]]

8. What Requirements Apply to Packaging and Label Operations? (Proposed 
Sec.  111.70)
    Proposed Sec.  111.70 would establish requirements for your 
packaging and label operations. The correct use of packaging and labels 
can affect whether your product is adulterated. For example, if a 
packaging material, intended only for use with a dry product, is used 
to package a liquid, unsafe substances could migrate from the packaging 
to the liquid, and adulterate your dietary ingredients or dietary 
supplements. In addition, if you apply the wrong label, your product 
would be adulterated under section 402(g) of the act because your label 
must be that which is specified in the master manufacturing record. In 
addition, your product would be misbranded under section 403 of the 
act.
    Proposed Sec.  111.70(a) would require that you take necessary 
actions to ensure each packaging container for holding dietary 
ingredients or dietary supplements meets its specifications so that the 
packaging container's condition will not contaminate your dietary 
ingredients or dietary supplements or cause them to deteriorate. As 
previously stated in the discussion of proposed Sec.  111.35(e)(4), you 
must establish specifications for packaging materials that may come in 
contact with dietary ingredients or dietary supplements. Meeting such 
specifications would ensure that the packaging that is used is safe and 
suitable for the intended use and meets all of the statutory and 
regulatory requirements under the act. In that way, the packaging 
materials will not adulterate the dietary ingredient or dietary 
supplement. This proposed requirement would give you the discretion to 
establish the specifications for each packaging container, and would 
require that these specifications are routinely met. For example, if 
your product is sensitive to light, you would choose a container that 
protects the product from the light so that it does not deteriorate.
    Proposed Sec.  111.70(b) would require that you fill, assemble, 
package, and perform other related operations in a way that protects 
your dietary ingredients or dietary supplements against adulteration 
and misbranding. The proposal would require that you use any effective 
means to do this, which would include:
    ? Cleaning and sanitizing all filling and packaging 
equipment, utensils, and dietary ingredient or dietary supplement 
containers, as appropriate. This is important because cleaning and 
sanitizing all filling and packaging equipment can help you avoid some 
common mistakes that can adulterate your products. For example, in one 
case, a consumer complained about receiving two different sized 
capsules in a bottle labeled as containing acidophilus capsules. We 
conducted an investigation and found that the manufacturer had received 
a similar report from a different consumer (Ref. 77). We analyzed the 
capsules and found that the smaller capsules were not acidophilus 
capsules but contained levels of stannous fluoride that would cause 
convulsions in certain persons and even exceeded the lethal dose in 
small children. We also collected unopened bottles of the acidophilus 
product and, after opening the product, found different sized capsules. 
The presence of smaller capsules containing stannous fluoride mixed in 
with the larger acidophilus capsules adulterated the product. The fact 
that these small stannous fluroride capsules mixed in with the larger 
acidophilus capsules indicated that the manufacturer had not cleaned 
the filling equipment properly.
    In another case, consumer complaints about a vitamin C product 
prompted us and the product's manufacturer to investigate the product 
(Ref. 78). We both discovered that the products contained niacin 
instead of vitamin C, and the problem was the result of a failure to 
clean out the packaging equipment so that niacin that had been left in 
the packaging equipment was put into the capsules during the 
manufacturing operation for the vitamin C product. The manufacturer 
reviewed its packing operations and instructed its personnel at the 
manufacturing plant to prevent this problem from reoccurring.
    ? Protecting manufactured dietary ingredients and dietary 
supplements from contamination, particularly airborne particulates such 
as dust, dirt, or microbes that may contaminate your product when your 
product is exposed to the environment.
    ? Using sanitary handling procedures.
    ? Establishing physical or spatial separation of packaging 
and labels from operations on other dietary ingredients and dietary 
supplements to prevent mixups. It is important to keep inprocess 
material separate from finished product that is ready to be packaged 
and labeled so that inprocess material is not inadvertently packaged 
and labeled as finished product. In addition, this proposed requirement 
would prevent mixup of one type of dietary ingredient with another type 
of dietary ingredient during packaging and label operations such as the 
vitamin C and niacin mixup described earlier.
    ? Identifying, by any effective means, filled dietary 
ingredient or dietary supplement containers that are set aside and held 
in unlabeled condition for future label operations, to prevent mixups;
    ? Identifying the dietary ingredient or dietary supplement 
with a batch, lot, or control number that can be used to determine the 
manufacturing history and control of the batch. Using a unique 
identifier for each batch or lot is necessary for you to trace the 
manufacturing history for a particular batch, and thus help you 
investigate and correct any safety problems for a batch or to recall a 
dietary ingredient or dietary supplement batch. For example, if you 
discovered a particular batch had a safety problem, you could recall 
the batch by identifying the batch number for the problem product. If 
you did not have a unique identifier, consumers would be unable to 
determine which product was the subject of a recall, and they may not 
stop using the product or you will have to recall more of the product.
    ? Examining a representative sample of the packaged and 
labeled dietary ingredient or dietary supplement to ensure that it 
meets specifications and that the label specified in the master 
manufacturing record has been applied; and
    ? Suitably disposing of labels and other packaging for 
dietary ingredients or dietary supplements that are obsolete or 
incorrect to ensure that they are not used in any future packaging and 
label operations. The use of any obsolete or incorrect label would 
adulterate the product because it would not comply with the requirement 
that the correct label as specified in the master manufacturing record 
be used.
    Proposed Sec.  111.70(c) would require that you conduct a material 
review and make a disposition decision of any packaged and labeled 
dietary ingredients or dietary supplements that do not meet 
specifications. If packaged and labeled dietary ingredients or dietary 
supplements do not meet specifications, it means that there is a 
problem and that the dietary ingredient or dietary supplement may be or 
is adulterated and this step is needed to determine what to do and how 
to handle the product to ensure that it does not get distributed.
    Sometimes problems arise because a manufacturer used the wrong 
label on a particular ingredient. For example, in one case, an 
ingredient manufacturer put the wrong label on its product so that a 
product labeled as containing zinc picolinate actually contained zinc 
polynicotinate (Ref. 79). The dietary

[[Page 12213]]

ingredient went to another manufacturer who, believing that the product 
was zinc picolinate, used the dietary ingredient to make its dietary 
supplement. The error was discovered after consumers who used the 
product started complaining of adverse reactions that are associated 
with niacin supplements, but the problem could have been avoided if the 
dietary ingredient manufacturer had taken steps to ensure that the 
correct labels were used.
    Proposed Sec.  111.70(d) would require that you repackage or 
relabel dietary ingredients or dietary supplements if approved and 
appropriately documented by your quality control unit. The quality 
control unit would need to decide whether the improperly packaged 
product was adulterated by the incorrect package and could be 
repackaged and relabeled without reprocessing of the dietary ingredient 
or dietary supplement.
    Proposed Sec.  111.70(e) would require that you retest or reexamine 
any repackaged or relabeled dietary ingredients or dietary supplements. 
They must meet all specifications and the quality control unit must 
approve or reject their release for distribution. The reason this is 
necessary is to ensure for example, by testing or examination, that the 
repackaged or relabeled product meets specifications and that the 
container in which the product is repackaged meets specifications.
    Proposed Sec.  111.70(f)(1) would require that you control the 
issuance and use of packaging and labels and reconcile the issuance and 
use of discrepancies. It is important to control access to the storage 
of packaging and labels; for example, only the labels that are required 
for current label operations should be issued to prevent issuance of 
any incorrect labels during the label operation. Using batch or lot 
numbers on your labels may be one control method. Batch or lot numbers 
also help you (and us) to identify a particular product and to trace 
that product's manufacturing history through your CGMP records. They 
can help identify which products are affected by a product recall, if a 
recall is necessary, and this can help preserve consumer confidence in 
your product.
    For example, if a recall covers batch A123, and a particular 
consumer has a product whose batch number is C456, he or she will know 
that the product is not covered by the recall. In contrast, if no batch 
numbers appear on the product label, the consumer would not be able to 
tell whether his or her product is covered by the recall and may 
continue to use it.
    As another example, controlling access of labels can help identify 
instances when mislabeling may have occurred. If you issue only the 
necessary number of labels to cover a particular production run but use 
fewer labels than expected even though you labeled the expected number 
of containers for the production run, this discrepancy would suggest 
that you used some wrong labels during the run and that you should 
conduct an investigation to determine the cause of, or reconcile the 
discrepancy.
    Proposed Sec.  111.70(f)(2) would require that you must examine 
carefully, before packaging operations, packaging and labels for each 
batch of dietary ingredient or dietary supplement to ensure that the 
label and packaging conform to the master manufacturing record.
    Proposed Sec.  111.70(g) would require that the person who performs 
the requirement established in accordance with this section document, 
at the time of performance, that he or she performed the requirement. 
This would include, but not be limited to, documentation in the batch 
production record of:
    ? The identity and quantity of the packaging and labels used 
and reconciliation of any discrepancies between issuance and use;
    ? The examination of a representative sample (as proposed 
Sec.  111.70(b)(7) would require);
    ? The conclusions you reached from retests conducted under 
proposed Sec.  111.70(e); and
    ? Any material reviews and disposition decisions for 
packaging and labels.
    Proposed Sec.  111.70(h) would require that you keep the packaging 
and label operations records required under this section established in 
accordance with proposed Sec.  111.125. These records are necessary to 
ensure that the correct packaging and label, i.e., the packaging and 
label specified by the master manufacturing record, were used in and 
applied to the batch of dietary ingredient or dietary supplement. These 
records together with the master manufacturing records and batch 
production records will ensure that a complete history of each batch of 
dietary ingredient or dietary supplement including use of the correct 
packaging and label is available for your review in the event that a 
problem arises with a particular batch. These records also are 
necessary to demonstrate compliance with the CGMP and quality control 
procedures. We invite comment on whether we should require, in a final 
rule, that you establish and follow written procedures for packaging 
and label operations that implement the requirements of this section. 
If comments assert that written procedures are necessary, comments 
should include an explanation of why the requirement is necessary to 
prevent adulteration including how such a requirement would ensure the 
identity, purity, quality, strength, and composition of the dietary 
ingredient or dietary supplement. Conversely, if comments assert that 
written procedures are not necessary, comments should include an 
explanation of why the requirement is not necessary including how, in 
the absence of the requirement, one can prevent adulteration and ensure 
the identity, purity, quality, strength and composition of the dietary 
ingredient or dietary supplement. Further, we seek comment on whether 
any of the proposed requirements in this section are not necessary to 
prevent adulteration and to ensure the identity, purity, quality, 
strength, and composition of the dietary ingredient or dietary 
supplement. If comments assert that certain provisions are not 
necessary, comments should include an explanation of why the 
requirement is not necessary including how, in the absence of the 
requirement, one can prevent adulteration and ensure the identity, 
purity, quality, strength, and composition of the dietary ingredient or 
dietary supplement. If comments agree that the proposed requirements 
are necessary for reasons other than those we have provided, the 
comments should so state and provide an explanation.
9. What Requirements Apply to Rejected Components, Dietary Ingredients, 
Dietary Supplements, Packaging, and Labels? (Proposed Sec.  111.74)
    Proposed Sec.  111.74 is intended to ensure that you do not 
mistakenly use rejected materials that are determined by the quality 
control unit to be unsuitable for use to make a dietary ingredient or 
dietary supplement.
    Proposed Sec.  111.74(a) would require that you clearly identify, 
hold, and control, under a quarantine system any component, dietary 
ingredient, dietary supplement, packaging, and label that is rejected 
and unsuitable for use in manufacturing, packaging, or label 
operations. The term ``control under a quarantine system'' indicates 
that you must prevent the use of any rejected component, dietary 
ingredient, dietary supplement, packaging, or label because such 
rejected product is unsuitable for use. For example, under this 
proposed rule, if a component, dietary ingredient, or dietary 
supplement is rejected and determined by the quality control unit

[[Page 12214]]

to be unsuitable for use, such material would be adulterated and not be 
suitable for reprocessing. Therefore, to prevent contamination of 
nonrejected material, you must quarantine the rejected material before 
disposal. The proposed rule would not specify any particular mechanism 
for how you quarantine the material, instead, you would have discretion 
in deciding what actions to take or what process to use.
    You also should note that, by referring to items that are rejected 
and unsuitable for use, proposed Sec.  111.74(a) excludes items that 
can be reprocessed and made suitable for use. Those items that can be 
reprocessed and made suitable for use are dealt with in proposed Sec.  
111.82.

F. Holding and Distributing (Proposed Subpart F)

1. What Requirements Apply to Holding Components, Dietary Ingredients, 
Dietary Supplements, Packaging, and Labels? (Proposed Sec.  111.80)
    Proposed Sec.  111.80 would require that you hold dietary 
ingredients and dietary supplements under conditions that will protect 
them against contamination and deterioration. Proposed Sec.  111.80(a) 
would require that you hold components, dietary ingredients, and 
dietary supplements under appropriate conditions of temperature, 
humidity, and light so that the identity, purity, quality, strength, 
and composition of the components, dietary ingredients, and dietary 
supplements are not affected. This proposed provision includes the 
holding of components, dietary ingredients, dietary supplements in your 
physical plant and at any point in the distribution process, however, 
we would not extend the holding requirements under this proposed CGMP 
regulation to retail establishments, but would defer to State and local 
governments for regulating operations that provide dietary supplements 
to retail for sale to the consumer. However, if a retail holding area 
is filthy, we would not be prevented from taking an enforcement action 
under a legal authority other than section 402(g) of the act.
    This requirement would ensure that products are not contaminated 
while they are held by the manufacturer, the wholesaler, or while being 
held at a warehouse. This would increase the likelihood that the 
products consumers purchase have the same quality as when they left the 
manufacturer. Note that proposed Sec.  111.80(a) uses the words ``not 
affected;'' this means that the conditions under which you hold 
components, dietary ingredients, and dietary supplements must not 
adulterate the components, dietary ingredients, or dietary supplements. 
For example, dried plants stored in a hot, humid warehouse may become 
moldy. Mold contamination could adversely affect the purity of the 
dietary ingredients and dietary supplements you manufacturer. You will 
decrease the chances of mold contaminating your dried plants if you 
control temperature and humidity.
    Proposed Sec.  111.80(b) would require that you hold packaging and 
labels under appropriate conditions of temperature, humidity, and light 
so that the quality of the packaging and labels are not affected. For 
example, some plastics become brittle when exposed to extreme 
temperatures. If brittle plastic containers are used to hold dietary 
ingredients or dietary supplements, they could crack or break, thereby 
losing their protective qualities, and lead to contamination or 
deterioration of the dietary ingredient or dietary supplement. You need 
to know the conditions of temperature, humidity, and light that are 
appropriate for your packaging and labels and you need to hold the 
packaging and labels under such conditions.
    Proposed Sec.  111.80(c) would require that you hold components, 
dietary ingredients, dietary supplements, packaging, and labels under 
conditions that do not lead to mixup, contamination, or deterioration 
of the components, dietary ingredients, dietary supplements, packaging, 
and labels. For example, your holding conditions must include a system 
for identifying container contents and its status (e.g., segregated, 
approved for use) in a manner that prevents mixup or use of unsuitable 
materials in manufacturing. Further, the presence of rodents in your 
holding area may cause contamination or deterioration of components, 
dietary ingredients, dietary supplements, packaging, and labels. 
Therefore, your holding conditions must be rodent-free. Moreover, 
rodents in your holding area would adulterate your dietary ingredient 
or dietary supplement under section 402(g) of the act. Holding 
conditions that prevent mixup, contamination, or deterioration of 
components, dietary ingredients, dietary supplements, packaging, or 
labels are necessary to prevent the production of an adulterated 
dietary ingredient or dietary supplement.
    We invite comment on whether we should require, in a final rule, 
that you establish and follow written procedures for holding 
components, dietary ingredients, dietary supplements, packaging, and 
labels. If comments assert that written procedures are necessary, 
comments should include an explanation of why the requirement is 
necessary to prevent adulteration including how such a requirement 
would ensure the identity, purity, quality, strength, and composition 
of the dietary ingredient or dietary supplement. Conversely, if 
comments assert that written procedures are not necessary, comments 
should include an explanation of why the requirement is not necessary 
including how, in the absence of the requirement, one can prevent 
adulteration and ensure the identity, purity, quality, strength, and 
composition of the dietary ingredient or dietary supplement.
2. What Requirements Apply to Holding Inprocess Material? (Proposed 
Sec.  111.82)
    Proposed Sec.  111.82 discusses proposed requirements for holding 
inprocess material. Proposed Sec.  111.82 would require that you 
segregate any inprocess material that does not meet your 
specifications, is awaiting further processing, or needs further 
evaluation by the quality control unit (e.g., because the inprocess 
material does not meet specifications, or because of an unexpected 
occurrence) to determine if it is suitable for reprocessing.
    Proposed Sec.  111.82(a), therefore, would require that you 
identify and hold inprocess material under conditions that will protect 
such material against mixup, contamination, and deterioration.
    Proposed Sec.  111.82(b) would require that you hold inprocess 
material under appropriate conditions of temperature, humidity, and 
light. The intent here is to prevent any contamination or deterioration 
of that inprocess material.
    We invite comment on whether we should require, in a final rule, 
that you establish and follow written procedures for holding inprocess 
material. If comments assert that written procedures are necessary, 
comments should include an explanation of why the requirement is 
necessary to prevent adulteration including how such a requirement 
would ensure the identity, purity, quality, strength, and composition 
of the dietary ingredient or dietary supplement. Conversely, if 
comments assert that written procedures are not necessary, comments 
should include an explanation of why the requirement is not necessary 
including how, in the absence of the requirement, one can prevent 
adulteration and ensure the identity, purity, quality, strength, and 
composition of the dietary ingredient or dietary supplement.

[[Page 12215]]

3. What Requirements Apply to Holding Reserve Samples of Components, 
Dietary Ingredients, and Dietary Supplements? (Proposed Sec.  111.83)
    Earlier, we discussed a provision concerning the collection of 
reserve samples. Proposed Sec.  111.50(h) would require that you 
collect representative reserve samples of each batch of dietary 
ingredient or dietary supplement. Proposed Sec.  111.83 would set forth 
requirements for holding any reserve samples collected.
    Proposed Sec.  111.83(a) would require that you hold any reserve 
samples of components or dietary ingredients collected in a manner that 
protects against contamination and deterioration.
    Proposed Sec.  111.83(b) would require that you hold such reserve 
samples of dietary supplements in a manner that protects against 
contamination and deterioration. Further, this provision would require 
that you hold the reserve samples under conditions of use recommended 
or suggested in the label of the dietary supplement and, if no 
conditions of use are recommended or suggested in the label, then under 
ordinary conditions of use. This proposed requirement also would 
require that you use the same container-closure system in which the 
dietary supplement is marketed or one that provides the same level of 
protection against contamination or deterioration as the marketed 
container-closure system. It is necessary to hold the reserve sample of 
a dietary supplement under the same conditions and in the same 
packaging as you would expect a consumer to hold that dietary 
supplement so that, if you need to later test that reserve sample, the 
testing would reflect current conditions under which the dietary 
supplement is held by the consumer prior to being consumed.
4. What Requirements Apply to Returned Dietary Ingredients or Dietary 
Supplements? (Proposed Sec.  111.85)
    Proposed Sec.  111.85 would establish requirements for returned 
dietary ingredients or dietary supplements. ``Returned'' dietary 
ingredients or dietary supplements are those products that a 
distributor, wholesaler, or retailer returns to a manufacturer. 
Proposed Sec.  111.85(a) would require that you identify returned 
dietary ingredients or dietary supplements and to quarantine them until 
your quality control unit conducts a material review and makes a 
disposition decision. (Your quality control unit would do this under 
proposed Sec.  111.37.) For example, you could attach a tag or other 
identifier on the returned dietary ingredient or dietary supplement to 
show that it is ``returned.'' We would require that you identify and 
quarantine (not just identify and segregate) returned dietary 
ingredients or dietary supplements so that they cannot be used. We 
propose to require that you quarantine returned products because you 
must assume that the returned product is adulterated until tests show 
otherwise. Thus, the product should not have physical closeness or 
contact with nonreturned product to ensure that it will not be mixed up 
mistakenly with nonreturned product, redistributed or reused in 
manufacturing.
    Proposed Sec.  111.85(b)(1) states that you may salvage returned 
dietary ingredients and dietary supplements only if:
    ? Evidence from their packaging (or, if possible, an 
inspection of the premises where the dietary ingredients and dietary 
supplements were held) indicates that the dietary ingredients and 
dietary supplements were not subjected to improper storage conditions. 
This would require that you have personal knowledge of the exact 
conditions under which the returned dietary ingredients or dietary 
supplements were held. Normally, for most types of packaging, simply 
examining the packaging will not tell you about the storage conditions 
that existed. However, we are aware of some technologies that are being 
used, such as temperature-sensitive materials that change colors, that 
could provide some information about storage conditions; and
    ? Tests demonstrate that the dietary ingredients or dietary 
supplements meet all specifications for identity, purity, quality, 
strength, and composition. This requirement will ensure that you do not 
return to distribution a dietary ingredient or dietary supplement that 
does not meet specifications. Salvage is available for only those 
products for which testing can be performed on finished product.
    For purposes of this discussion, ``salvage'' means to return to 
distribution without reprocessing the dietary ingredient or dietary 
supplement.
    Proposed Sec.  111.85(c) would require that you destroy or suitably 
dispose of the returned dietary ingredients or dietary supplements if 
they do not meet specifications for identity, purity, quality, 
strength, and composition, unless the quality control unit conducts a 
material review and makes a disposition decision to allow reprocessing.
    Proposed Sec.  111.85(d) would require that you conduct an 
investigation of your manufacturing processes and those other batches 
if the reason for a dietary ingredient or a dietary supplement being 
returned implicates other batches. The point of the investigation would 
be to determine whether, for example, the other implicated batches may 
have the same problem or have been subject to the same problematic 
manufacturing process for which the dietary ingredient or dietary 
supplement was returned. Other batches may be implicated if the 
component or dietary ingredient used in the returned product also was 
used in additional batches or if your investigation indicates that 
there was a problem with a step in the manufacturing process that 
affected additional batches. The proposal also would require that you 
document the investigation and include your conclusions and followup.
    Proposed Sec.  111.85(e) would require you to establish and keep 
records for any material review and disposition decision and any 
required testing to determine compliance with specifications done for 
any returned dietary ingredient or dietary supplement. You should 
include the following information in your records:
    ? The name of the person or company or both the name of the 
person and company who returned the dietary ingredients or dietary 
supplements;
    ? A description of the returned dietary ingredient or dietary 
supplement;
    ? The batch or lot number of the returned dietary ingredient 
or dietary supplement and any reprocessed batch or batch manufactured 
using the returned dietary ingredient or dietary supplement;
    ? The reason for the return;
    ? The quantity returned;
    ? The disposition of the dietary ingredient or dietary 
supplement; and
    ? The date of disposition.
    Proposed Sec.  111.85(f) would require that you make and keep 
records for returned dietary ingredients and dietary supplements in 
accordance with Sec.  111.125. These records are necessary to ensure 
that returned products that could be adulterated are not inadvertently 
redistributed or inadvertently used in manufacturing. Further, records 
of any reprocessed batch or batch manufactured using the returned 
product will be useful in the event that a problem arises with a 
particular batch that is manufactured with returned product. These 
records also are necessary to demonstrate compliance with the CGMP and 
quality control procedures. We invite comment on whether we should 
require, in a final rule, that you establish and follow

[[Page 12216]]

written procedures for identifying, quarantining, and salvaging 
returned dietary ingredients and dietary supplements. If comments 
assert that written procedures are necessary, comments should include 
an explanation of why the requirement is necessary to prevent 
adulteration including how such a requirement would ensure the 
identity, purity, quality, strength, and composition of the dietary 
ingredient or dietary supplement. Conversely, if comments assert that 
written procedures are not necessary, comments should include an 
explanation of why the requirement is not necessary including how, in 
the absence of the requirement, one can prevent adulteration and ensure 
the identity, purity, quality, strength, and composition of the dietary 
ingredient or dietary supplement.
5. What Requirements Apply to Distributing Dietary Ingredients or 
Dietary Supplements? (Proposed Sec.  111.90)
    Proposed Sec.  111.90 would establish requirements concerning the 
distribution of dietary ingredients and dietary supplements. Proposed 
Sec.  111.90(a) would require any distribution of dietary ingredients 
or dietary supplements to be under conditions that will protect them 
from contamination and deterioration. This is to protect dietary 
ingredients and dietary supplements from distribution practices that 
may adulterate them.
    As discussed previously, proposed part 111 also would apply to 
foreign firms that manufacture, package, or hold dietary ingredients 
and dietary supplements that are imported or offered for import into 
the United States, unless imported for further processing and export 
under section 801(d)(3) of the act. It also would apply to persons who 
distribute imported dietary ingredients and dietary supplements, and to 
persons who export dietary ingredients and dietary supplements from the 
United States unless exported in compliance with section 801(e) of the 
act.
    We recognize that the safety of dietary supplements cannot be 
adequately ensured if the imports are not subject to the same controls 
as domestic products. In addition, we believe that the importer who 
distributes a foreign product should share responsibility with the 
foreign manufacturer for safety. More often than not, it is a U.S. 
importer, rather than the foreign manufacturer, who actually 
distributes imported dietary supplements for sale in the United States. 
Thus, we believe that importers of dietary ingredients or dietary 
supplements should take steps to ensure that their shipments are 
obtained from manufacturers that follow these proposed CGMP 
requirements.
    In addition, these proposed CGMPs would apply to manufacturers who 
export their dietary ingredient or dietary supplement, unless exported 
in compliance with section 801(e) of the act. Section 801(e)(1) of the 
act states that a food intended for export must not be deemed to be 
adulterated or misbranded under the act if it:
    ? Accords to the foreign purchaser's specifications;
    ? Is not in conflict with the laws of the country to which it 
is intended for export;
    ? Is labeled on the outside of the shipping package that it 
is intended for export; and
    ? Is not sold or offered for sale in domestic commerce.
    Dietary ingredients and dietary supplements for export are subject 
to section 801(e)(1) of the act and would be subject to the 
notification and recordkeeping requirements of Sec.  1.101 (21 CFR 
1.101) and you would be required to comply with the export requirements 
of Sec.  1.101.
    We invite comment on whether we should require, in a final rule, 
that you make and keep records on the distribution of dietary 
ingredients and dietary supplements that you manufacture, package, or 
hold.

G. Consumer Complaints--What Requirements Apply to Consumer Complaints? 
(Proposed Subpart G, Sec.  111.95)

    Proposed Sec.  111.95 would establish requirements for receiving 
and handling consumer complaints. Consumer complaints can be helpful in 
alerting you to possible manufacturing and safety problems associated 
with your dietary ingredients or dietary supplements.
    As stated in Sec.  111.3, consumer complaint refers to a possible 
failure of a dietary ingredient or dietary supplement to meet any of 
the requirements of this part, including those that, if not met, may 
result in a possible risk of illness or injury. Proposed Sec.  
111.95(e) would require that you keep a written record of every 
consumer complaint that is related to good manufacturing practices. 
Thus, whether the complaint was sent by regular mail, electronic mail, 
or any other form of written communication, or whether received orally, 
you would be required to keep a written record of each consumer 
complaint. You should include all information that would allow your 
quality control unit to determine whether an investigation of the 
complaint is necessary.
    Proposed Sec.  111.95(a) would require that you have a qualified 
person review all consumer complaints to determine whether the consumer 
complaint involves a possible failure of a dietary ingredient or 
dietary supplement to meet any of its specifications, or any other 
requirements of this part, including those specifications and other 
requirements that, if not met, may result in a possible risk of illness 
or injury. A ``qualified person'' would be a person who has the 
training and experience to determine whether a complaint represents a 
possible failure of a dietary ingredient or dietary supplement to meet 
any of the requirements in this part, or represents a possible risk of 
illness or injury that is unrelated to such failure. The qualified 
person's review is important for distinguishing between those consumer 
complaints that your quality control unit must review and those 
consumer complaints that represent a consumer's dissatisfaction with a 
dietary ingredient or dietary supplement that is unrelated to a 
possible failure to meet specifications that would be required by this 
proposal, or any other requirement in this part. For example, some 
consumer complaints about quality may simply express a personal dislike 
of the taste, color, odor, or size of tablet, which would probably not 
require your quality control unit to review them. As stated earlier, 
consumer complaints related to an illness or injury related to a 
pharmacologically active substance of a particular dietary ingredient, 
such as aristolochic acid, would not be a consumer complaint within the 
meaning of that term in this proposal and thus would not be of the type 
that the quality control unit must review under this proposed rule.
    Proposed Sec.  111.95(b) would require that your quality control 
unit review all consumer complaints involving the possible failure of a 
dietary ingredient or dietary supplement to meet any of its 
specifications, or any of the other requirements in this part, 
including those specifications and other requirements that, if not met, 
may result in a possible risk of illness or injury, to determine 
whether there is a need to investigate the consumer complaint. When 
there is a reasonable possibility of a relationship between the quality 
of a dietary supplement and an adverse event, such as a report of an 
illness or injury that may be due to a wrong ingredient or wrong label, 
then the manufacturer would be required to do an investigation that 
includes both batch records associated with the

[[Page 12217]]

dietary ingredient or dietary supplement involved in the consumer 
complaint. However, if the quality control unit determines that an 
investigation is unnecessary, it would be helpful to you if your 
quality control unit documents why an investigation was not necessary. 
This information would be useful to you because it could save time if 
you receive additional similar consumer complaints about a particular 
product.
    Proposed Sec.  111.95(c) would require that your quality control 
unit investigate a consumer complaint when there is a reasonable 
possibility of a relationship between the quality of a dietary 
supplement and an adverse event. For example, if a manufacturer uses 
too much of a dietary ingredient in a dietary supplement (e.g., 400 to 
4,699 [mu]g of selenium instead of 200 [mu]g of selenium), it is a 
manufacturing error that may result in an adverse event. Further, if a 
communication alleges consumer dizziness, vomiting, or lightheadedness 
after consuming several dietary supplements, it is a adverse event 
report that is worthy of quality control unit investigation.
    Proposed Sec.  111.95(d) would describe what the quality control 
unit's investigation must include. In brief, the quality control unit's 
investigation of a consumer complaint must include the batch records 
associated with the dietary ingredient or dietary supplement involved 
in the consumer complaint. The quality control unit must extend the 
investigation to other batches of dietary ingredients or dietary 
supplements that may have been associated with a failure to meet a 
specification or any other requirements of this part. When there is a 
possible product defect or failure, we recommend that the investigation 
include laboratory testing of the dietary ingredient or dietary 
supplement because you will need the test results to determine if 
specifications or requirements for the dietary ingredient or dietary 
supplement were not met. Complaints such as those that involve serious 
adverse events should include followup by a health care provider. For 
other types of complaints, neither laboratory nor medical investigation 
may be necessary because the product defect or failure may be 
identified by reviewing batch documents or the consumer complaint may 
not involve a serious adverse event.
    Proposed Sec.  111.95(e) would require that you make and keep a 
written record of every consumer complaint that is related to good 
manufacturing practices. For the purposes of this regulation, a 
consumer complaint about product quality may or may not include 
concerns about a possible hazard to health. However, a consumer 
complaint does not include an adverse event, illness, or injury related 
to the safety of a particular dietary ingredient independent of whether 
the product is produced under good manufacturing practices. The 
consumer complaint written record must include, but is not limited to, 
the following:
    ? The name and description of the dietary ingredient or 
dietary supplement;
    ? The batch or lot number of the dietary supplement, if 
available;
    ? The complainant's name, if available;
    ? The nature of the complaint, including how the consumer 
used the product;
    ? The reply to the complainant, if any; and
    ? Findings of the investigation and followup action taken 
when an investigation is performed.
    We suggest that you report the consumer complaint and the 
investigation results to us when there is a possibility of a 
relationship between the consumption of a dietary supplement and a 
serious adverse event. While the proposal would not require that you 
submit these reports, we strongly suggest that you do so because we may 
have additional expertise or data that may be helpful in investigating 
the complaint or determining whether the problem applies to more than 
one product. We suggest that you submit these reports within 15 days 
after you receive such information to the FDA MedWatch program by 
calling our ``MedWatch'' program (our database for reporting possible 
adverse events) at 1-800-FDA-1088 (1-800-332-1088) to request that a 
reporting form (one-page, return postage paid) and instructions on how 
to complete the form be mailed to you, downloading a form and 
instructions from the MedWatch Internet site at http://www.fda.gov, Exit Disclaimer or 
using the interactive form available on the MedWatch Internet site at 
http://www.fda.gov. Exit Disclaimer
    Further, we suggest that you report a consumer complaint even if 
you are not the manufacturer of a dietary ingredient or dietary 
supplement and only package or distribute a dietary ingredient or 
dietary supplement if you receive a consumer complaint that may be 
related to the manufacture of the dietary ingredient or dietary 
supplement. Sometimes consumers submit complaints to the person who 
distributes a product or the person who is listed on the package label. 
If this happens, you should notify the manufacturer of the dietary 
ingredient or dietary supplement of the consumer complaint because the 
manufacturer may not be aware of possible problems associated with its 
products.
    Proposed Sec.  111.95(f) addresses documentation and recordkeeping. 
Consumer complaints can alert you (and us) to potential quality 
problems with a product that is related to good manufacturing 
practices, such as cases where the manufacturer used the wrong 
ingredient or put the wrong label on a product. A prudent manufacturer, 
therefore, must investigate any complaints regarding its products 
because the results of its investigations might lead to solutions or 
improvements that will make the product or manufacturing process better 
and benefit the manufacturer and consumers.
    Proposed Sec.  111.95(f)(1) would require the person who performs 
the requirement established in accordance with this section to 
document, at the time of performance, that he or she performed the 
requirement.
    Finally, proposed Sec.  111.95(f)(2) would require that you keep 
consumer complaint records established in accordance with proposed 
Sec.  111.125. These records are necessary for handling consumer 
complaints in a manner that ensures that an unanticipated problem with 
a dietary ingredient or dietary supplement is reviewed and 
investigated. These records also are necessary to demonstrate 
compliance with the CGMP.
    We invite comment on whether we should require, in a final rule, 
that you establish and follow a written procedure for receiving, 
reviewing, and investigating consumer complaints. If comments assert 
that written procedures are necessary, comments should include an 
explanation of why the requirement is necessary to prevent adulteration 
including how such a requirement would ensure the identity, purity, 
quality, strength, and composition of the dietary ingredient or dietary 
supplement. Conversely, if comments assert that written procedures are 
not necessary, comments should include an explanation of why the 
requirement is not necessary including how, in the absence of the 
requirement, one can prevent adulteration and ensure the identity, 
purity, quality, strength, and composition of the dietary ingredient or 
dietary supplement.

H. Records and Recordkeeping--What Requirements Apply to Recordkeeping? 
(Proposed Subpart H, Sec.  111.125)

    Throughout this discussion of the proposed rule, some provisions 
have included a paragraph that would require

[[Page 12218]]

that you keep records established in accordance with proposed Sec.  
111.125. Proposed Sec.  111.125 would establish general recordkeeping 
requirements and tell you how long you must keep certain records. As we 
have stated several times in this document, we determine CGMP 
compliance by conducting inspections. Records, therefore, enable you to 
show, and for us to determine, how you complied with the CGMP 
requirements.
    Proposed Sec.  111.125(a) would apply to all records covered by the 
proposed rule and would require that you keep those records for 3 years 
beyond the date of manufacture of the last batch of dietary ingredients 
or dietary supplements associated with those records. Retention for 3 
years beyond the date of manufacture would be appropriate for followup 
of consumer complaints received during the marketing period.
    Proposed Sec.  111.125(b) would deal with the form in which you 
keep records. The proposal would allow you to keep records required 
under this part as original records, as true copies (such as 
photocopies, microfilm, microfiche, or other accurate reproductions of 
the original records), or as electronic records. If you use reduction 
techniques, such as microfilming, the proposal would require that you 
make suitable reader and photocopying equipment readily available to 
us. If you use electronic records, the proposal would require that you 
comply with part 11 (our requirements for electronic records).
    Proposed Sec.  111.125(c) would require that you make your records 
available for inspection and copying by us when requested. We sometimes 
need to copy records when our field inspectors need guidance or 
additional expertise from our headquarters staff; if we were unable to 
copy records, our inspections would become more complicated and longer 
in duration, particularly if the inspection involved a complex 
scientific or technical issue that normally would be handled at FDA 
headquarters.

IV. Statement Concerning the Use of Plain Language

    In response to the June 1, 1998, White House Presidential 
Memorandum on Plain Language, we drafted this proposed rule in plain 
language. Plain language is intended to help readers find requirements 
quickly and understand them easily. To do that, we have reorganized 
sections modeled after existing regulations and reworded the paragraphs 
using:
    ? Short sections, paragraphs, sentences, and words to speed 
up reading and enhance understanding;
    ? Sections as questions and answers to focus sections better; 
and
    ? Personal pronouns to reduce passive voice and draw readers 
into the text.
    In some cases, we modeled a proposed provision after an existing 
regulation, but wrote the proposed rule using plain language 
techniques. We invite the public to comment on the plain language 
techniques used in this proposed rule. In developing your comments, 
please consider addressing the following points:
    ? Do you like the proposed rule's appearance?
    ? Do plain language techniques make the document easier to 
read and understand? and
    ? Do you have other suggestions to improve the format?

V. Paperwork Reduction Act of 1995

    This proposed rule contains information collection requirements 
that are subject to review by the Office of Management and Budget (OMB) 
under the Paperwork Reduction Act of 1995 (44 U.S.C. 3501-3520). A 
description of these requirements is given below with an estimate of 
the annual recordkeeping burden. Included in the estimate is the time 
for reviewing instructions, searching existing data sources, gathering 
and maintaining the data needed, and completing and reviewing each 
collection of information.
    We invite comments on: (1) Whether the proposed collection of 
information is necessary for the proper performance of our functions, 
including whether the information will have practical utility; (2) the 
accuracy of our estimate of the burden of the proposed collection of 
information, including the validity of the methodology and assumptions 
used; (3) ways to enhance the quality, utility, and clarity of the 
information to be collected; and (4) ways to minimize the burden of the 
collection of information on respondents, including the use of 
automated collection techniques, when appropriate, and other forms of 
information technology.
    Title: Current Good Manufacturing Practice in Recordkeeping and 
Reporting for Dietary Ingredients and Dietary Supplements.
    Description: Section 402(g) of the act gives us explicit authority 
to issue a rule regulating conditions for manufacturing, packaging, and 
holding dietary supplements. Section 402(g)(1) of the act states that a 
dietary supplement is adulterated if ``it has been prepared, packed, or 
held under conditions that do not meet current good manufacturing 
practice regulations.'' Section 402(g)(2) of the act authorizes us to, 
by regulation, ``prescribe good manufacturing practices for dietary 
supplements.'' Other relevant legal authority is discussed in section 
II of this document.
    For this proposed CGMP rule for dietary ingredients and dietary 
supplements, recordkeeping is necessary to provide the type of 
documentation that would demonstrate that dietary ingredients and 
dietary supplements are manufactured, packaged, and held under the 
conditions that would be required under the proposed CGMP regulations. 
Under section 701(a) of the act, we may issue regulations necessary for 
the efficient enforcement of the act. If you did not keep records, for 
example, documenting practices performed during previous production 
runs, it would be difficult for us to determine whether, as stated 
under section 402(g)(1) of the act, the dietary supplement had been 
manufactured, packaged, and held under CGMPs. By requiring records, we 
will be able to ensure that you follow CGMPs and that your dietary 
supplements are not adulterated and misbranded during manufacturing, 
packaging, or holding operations.
    The proposed rule would establish the minimum manufacturing 
practices necessary to ensure that dietary supplements are 
manufactured, packaged, or held in a manner that will not adulterate 
and misbrand the dietary ingredients or dietary supplements.
    The proposed regulations would impose requirements for: (1) 
Personnel, (2) physical plants, (3) equipment and utensils, (4) 
production and process controls, (5) holding and distributing, (6) 
consumer complaints, and (7) records and recordkeeping.
    We are proposing recordkeeping requirements that include records 
pertaining to: (1) Calibration of instruments and controls; (2) 
automatic, mechanical, or electronic equipment calibration, inspection, 
or checks; (3) production and process controls; (4) quality control; 
(5) receiving components, dietary supplements, packaging, and labels; 
(6) master manufacturing and batch production; (7) packaging and label 
operations; (8) returned dietary ingredients or dietary supplements; 
and (9) consumer complaints.
    Description of Respondents: Dietary ingredient manufacturers, 
dietary supplement manufacturers, packagers and repackagers, 
distributors, warehousers, exporters, importers, large businesses, and 
small businesses.
    We estimate the burden of this collection of information as 
follows:

[[Page 12219]]

                               Table 1.--Estimated Annual Recordkeeping Burden \1\
----------------------------------------------------------------------------------------------------------------
                                                      Annual
         21 CFR Section              Number of     frequency of    Total annual      Hours per      Total hours
                                   recordkeepers   recordkeeping      records         record
----------------------------------------------------------------------------------------------------------------
111.15(b)(3)....................             231              12           2,772            0.1              277
111.15(d)(3)....................             231             260          60,060            0.25          15,015
111.25(d).......................             213             365          77,745            0.5           38,873
111.30(b)(2) and (b)(5).........             707             260         183,820            0.5           91,910
111.35(d).......................              10               1              10           10                100
111.35(e).......................             367             260          95,420            0.25          23,855
111.35(f).......................             367             260          95,420            0.1            9,542
111.35(i)(1)....................             367              10           3,670            0.25             918
111.35(j).......................             367             260          95,420             .25          23,855
111.35(m).......................             367             365         133,955            0.1           13,396
111.37(b)(1), (b)(3) through                 286             260          74,360            0.5           37,180
 (b)(5), (b)(7) through (b)(10),
 and (b)(12)(i).................
111.37(c).......................             286             365         104,390            0.5           52,195
111.40(a)(3), (a)(4), (b)(2),                449             365         163,885            0.1           16,389
 and (b)(3).....................
111.40(c)(1)....................             218             365          79,570            0.5           39,785
111.45(a) \2\ and (b) \2\.......             200               1             200           30              6,000
111.50(a) through (c), (d)(1),                68             260          17,680            1             17,680
 and (e)........................
111.50(g).......................              68             260          17,680            0.5            8,840
111.60(b)(2)....................             133             365          48,545            1             48,545
111.60(d) \2\...................             133               1             133            3                399
111.65(c)(7), (c)(10), and                   133             365          48,545            0.1            4,855
 (c)(11)........................
111.70(b)(5) through (b)(6),                 245             260          63,700            0.1            6,370
 (d), and (e)...................
111.70(g).......................             245             260          63,700            0.50          31,850
111.74(a).......................             200              12           2,400            0.1              240
111.82(a).......................              53              52           2,756            0.1              276
111.85(a).......................              53             260          13,780            0.1            1,378
111.85(d) and (e)...............              53             260          13,780            0.5            6,890
111.95(e).......................              53              75           3,975            0.1              398
111.95(f)(1)....................              93              75           6,975            0.5            3,488
111.125.........................             220               4             880            0.1               88
                                                                                                 ---------------
    Total.......................  ..............  ..............  ..............  ..............        500,587
----------------------------------------------------------------------------------------------------------------
\1\ There are no capital costs or operating and maintenance costs associated with this collection of
  information.
\2\ One time burden.

    The burden estimates above are based on our institutional 
experience with CGMP requirements for drugs and on data provided by 
Research Triangle Institute (RTI) in the ``Survey of Manufacturing 
Practices in the Dietary Supplement Industry'' (Refs. E1 and E2). We 
tentatively conclude that there are no capital costs or operating costs 
associated with this proposed rule. However, we invite comments on 
information provided in table 1 of this document or on any anticipated 
costs.
    The estimates for number of firms affected by each provision of the 
rule are based on the percentage of manufacturers, ingredient 
suppliers, repacker/relabelers, distributors, and warehousers that 
reported to RTI that they have not established or do not maintain 
records that would be required or recommended under the proposed rule. 
The RTI survey estimated that 1,566 firms would be covered by this rule 
including manufacturers, dietary ingredient suppliers, repacker/
relabelers, distributors, and warehousers. The time estimates include 
the burden involved in documenting that certain requirements are 
performed and in recordkeeping. We used an estimated annual batch 
production of 260 batches per year to estimate the burden of 
requirements that are related to the number of batches produced 
annually, e.g., proposed Sec.  111.50, ``What requirements apply to 
establishing a batch production record?'' The estimate of 260 batches 
per year is near the midpoint of the number of annual batches reported 
by RTI survey firms.
    Proposed Sec.  111.125 prescribes the length of time for which CGMP 
records must be maintained. The burden chart reflects the estimated 
annual burden for record maintenance, for periodically reviewing 
records to determine if they may be discarded, and for any associated 
documentation for that activity for records that would be required 
under part 111. To avoid double-counting, we have not included a 
separate estimate of burden for those sections that would require 
maintaining records in accordance with proposed Sec.  111.125, but have 
included a single burden estimate for all such records maintenance 
under proposed Sec.  111.125. For example, proposed Sec.  111.50(a) 
would require that the batch production records be prepared every time 
a batch is manufactured and Sec.  111.50(i) would require that batch 
production records be kept in accordance with proposed Sec.  111.125. 
The estimated burden for establishing the batch production records is 
counted in proposed Sec.  111.50(a) and the estimated burden for 
keeping the batch production records as would be required in accordance 
with Sec.  111.50(i) is counted in proposed Sec.  111.125.
    In compliance with the Paperwork Reduction Act of 1995 (44 U.S.C. 
3507(d)), the agency has submitted a copy of this proposed rule to OMB 
for its review. Interested persons are requested to send comments 
regarding information collection to the Office of Information and 
Regulatory Affairs, OMB (see ADDRESSES).

VI. Environmental Impact Considerations

    The agency has determined under 21 CFR 25.30(j) that this action is 
of a type that does not individually or cumulatively have a significant 
effect on the human environment. Therefore,

[[Page 12220]]

neither an environmental assessment nor an environmental impact 
statement is required.

VII. Analysis of Impacts

A. Introduction

    FDA has examined the economic implications of this proposed rule as 
required by Executive Order 12866. Executive Order 12866 directs 
agencies to assess all costs and benefits of available regulatory 
alternatives and, when regulation is necessary, to select regulatory 
approaches that maximize net benefits (including potential economic, 
environmental, public health and safety, and other advantages; 
distributive impacts; and equity). Executive Order 12866 classifies a 
rule as significant if it meets anyone of a number of specified 
conditions, including: Having an annual effect on the economy of $100 
million, adversely affecting a sector of the economy in a material way, 
adversely affecting competition, or adversely affecting jobs. A 
regulation is also considered a significant regulatory action if it 
raises novel legal or policy issues. FDA has determined that this 
proposed rule, if it were to become a final rule, would be a 
significant regulatory action as defined by Executive Order 12866.
    The Small Business Regulatory Enforcement Fairness Act of 1996 
(Pub. L. 104-121) defines a major rule for the purpose of congressional 
review as being likely to cause one or more of the following: an annual 
effect on the economy of $100 million; a major increase in costs or 
prices; significant adverse effects on competition, employment, 
productivity, or innovation; or significant adverse effects on the 
ability of U. S.-based enterprises to compete with foreign-based 
enterprises in domestic or export markets. In accordance with the Small 
Business Regulatory Enforcement Fairness Act, OMB has determined that 
this proposed rule, when final, will be a major rule for the purpose of 
congressional review.
    FDA has examined the economic implications of this proposed rule as 
required by the Regulatory Flexibility Act (5 U.S.C. 601-612). If a 
rule has a significant economic impact on a substantial number of small 
entities, the Regulatory Flexibility Act requires agencies to analyze 
regulatory options that would lessen the economic effect of the rule on 
small entities. FDA finds that this proposed rule would have a 
significant economic impact on a substantial number of small entities.
    We carry out the cost-benefit analyses required for significant 
rules in the Preliminary Regulatory Impact Analysis, in section VII.B 
of this document. We perform the Initial Regulatory Flexibility 
Analysis of the effects on the proposed rule on small businesses in 
section VII.C of this document.

B. Preliminary Regulatory Impact Analysis

1. The Need for the Proposed CGMP Regulations
    The proposed CGMP regulations are needed because establishments 
that manufacture, package, and hold dietary ingredients and dietary 
supplements may not have sufficient market incentives to use controls 
to prevent the adulteration and misbranding of dietary ingredients or 
dietary supplements, including incentives to ensure their identity, 
purity, quality, strength, and composition (product quality). 
Manufacturing, packaging, and holding practices that ensure product 
quality can be costly, so establishments may not adopt them unless 
required to do so by regulation. Without the proposed regulations 
consumers of dietary supplements cannot be assured that all 
establishments are manufacturing dietary supplements in a way that 
ensures that these products are not adulterated or misbranded.
    Manufacturing, packaging, and holding practices can compromise 
safety if they fail to prevent biological, chemical, and physical 
contamination, or if the wrong dietary ingredients are used that 
present an unreasonable risk of illness or injury. Strength (which is 
the amount of a specific dietary supplement or dietary ingredient in 
each tablet or capsule) that differs from label statements, missing or 
extra ingredients, and inconsistency across units of the product are 
other problems caused by poor manufacturing practices. Products may 
also be held in insanitary or environmentally inappropriate conditions, 
or may be physically damaged if stored improperly. Some poor 
manufacturing practices, such as the use of ingredients that are 
undeclared, of incorrect strength, or missing altogether result in a 
misbranded product. The proposed CGMP regulations would establish 
minimum requirements to ensure that manufacturing, packaging, and 
holding practices ensure the identity and quality of components, 
dietary ingredients, and dietary supplements.
    Consumers today rely on manufacturer's assurances, existing 
regulations and statutes (for example, section 402(a)(3) and (a)(4) of 
the act), and recourse to the legal system to ensure that products are 
not defective. Brand names convey some information to consumers about a 
firm's manufacturing practices. Some private organizations, such as the 
National Nutritional Foods Association and the USP design minimum 
product standards or manufacturing requirements. The current act 
contains some provisions that prevent using putrid substances and 
insanitary manufacturing practices. In addition, either the threat of 
litigation or consumers seeking compensation for defective products and 
adverse health events may create incentives for establishments to adopt 
good manufacturing practices.
    Actions by manufacturers, primarily voluntary quality controls, do 
not provide sufficiently protective industry-wide minimum requirements 
for manufacturing, packaging, and holding of dietary ingredients and 
dietary supplements. Without the proposed regulations, survey evidence 
shows that products in the dietary supplement market are sorted 
somewhere between two types:
    ? Higher-priced products with brand names or industry 
certification that follow several of the good manufacturing practices 
proposed here;
    ? Lower-priced products that contain no private certification 
or respected brand name and that follow few of the good manufacturing 
practices that are proposed here.
    Without the proposed rule, the current practices do not provide all 
consumers with safe manufacturing practices or reliable product quality 
throughout the industry.
    The market for dietary supplements is full of information; 
consumers of dietary supplements must sort through information and 
misinformation about the properties of these products from magazines, 
brochures, popular books, television, and a host of other sources. 
However, the information from these sources deals most often with the 
claims for the products themselves, not with the steps taken by 
establishments to protect against contamination or to ensure quality. 
Private quality control fails to provide industry-wide minimum good 
manufacturing practices for the following reasons:
    ? Establishments do not have incentives to disclose 
information about their own practices, because disclosure that some 
consumers may perceive to be harmful or undesirable would reduce the 
demand for their products. Establishments therefore have incentives to 
withhold information from consumers.
    ? Businesses normally do not advertise differences in 
manufacturing

[[Page 12221]]

practices. They seldom have access to competitors' proprietary 
information, and they may fear that advertising based on differences in 
practices would discredit the entire industry.
    ? Without public disclosure of product quality and adverse 
health events, the link between manufacturing practice and health 
hazard is difficult to establish. The link is probabilistic, requires 
data pooling across products and establishments (in order to establish 
cross sectional variation), and can be interpreted in a variety of 
ways.
    ? Because many consumers already mistakenly believe that the 
Federal Government guarantees safety, businesses have weak incentives 
to adopt good manufacturing practices, which are costly. In one recent 
survey of the nation's consumers, 34 percent report that they believe 
that the government regulates dietary supplements to ensure safety and 
that products do what they claim to do. (For details of the survey, see 
Ref. E3.) If people believe that good manufacturing practices are 
already followed, manufacturers may believe they gain little from 
voluntarily adopting them.
    Information about manufacturing practices for dietary supplements 
is imperfect and costly to produce, so well-informed people should be 
willing to pay for improvements in the quality of information. An 
important benefit of the proposed regulations will be to reduce 
variation in manufacturing practices and ensure minimum quality for 
dietary supplement products. Reducing the variation in product quality 
by creating industry-wide minimum requirements reduces the information 
consumers now attempt to get through costly and uncertain sources in 
order to make purchasing decisions.
2. Regulatory Options
    FDA considered several regulatory options for dealing with current 
manufacturing, packaging, and holding practices that may not ensure 
product quality. The options considered include: (a) No new regulatory 
action, (b) fewer requirements for vitamins and minerals, (c) more 
restrictive regulations than the proposed CGMP regulations, (d) HACCP 
without the other elements of CGMP regulations, (e) final product 
testing only, (f) regulations for high-risk products or hazards only, 
and (g) the proposed rule.
    a. No new regulatory action. Under this option, consumers would 
probably rely on the following as protection against defective 
products:
    ? Possible enforcement action by FDA under, for example, 
section 402(a)(3) and (a)(4) of the act, regarding adulterated foods 
that consist of filthy, putrid, or decomposed substances or foods that 
have been prepared, packed, or held under insanitary conditions so that 
they may become contaminated or may be rendered injurious to health;
    ? Publicity from private consumer groups or health agencies 
on the risks from products not manufactured using CGMP regulations, 
manufacturers assurances, and the voluntary adoption of some or all 
provisions of the proposed regulations;
    ? Current or enhanced State and local enforcement activity to 
bring about a reduction of potential harm from contaminated or poor 
quality dietary supplements; or
    ? Litigation or the threat of litigation by consumers who 
allege harm from consumption of the dietary supplement.
    We believe that there are compelling reasons not to rely on these 
alternatives alone.
    If public and private health agencies, consumer groups, 
competitors, trade organizations or other third parties publicized the 
risks from products not manufactured using private good manufacturing 
practices, then consumers would decide for themselves on the risks of 
contaminated or poor quality products. The weakness of this alternative 
is that third-party organizations cannot easily discover many of the 
problems caused by poor manufacturing practices because manufacturers 
are reluctant to voluntarily share information to third parties about 
their manufacturing practices.
    Actions by manufacturers, such as by voluntarily introducing good 
manufacturing practices, occur when the expected private economic 
benefits of the actions exceed the private costs. Voluntary adoption of 
good manufacturing practices will occur when it is profitable to do so. 
Many establishments appear to be adopting some publicly available good 
manufacturing practice models in order to meet the demand for safer and 
more uniform products. NNFA is implementing a good manufacturing 
practice certification program. The USP sets standards for strength, 
purity, disintegration, and dissolution for individual and combination 
vitamins and minerals. Also, Consumerlab.com is introducing a 
certification label, CL, to show when ingredients meet their minimum 
requirements. However, 36 percent of recently surveyed dietary 
supplement establishments do not follow any good manufacturing practice 
models for their products (Ref. E2). The breakdown of survey results 
shows that 48 percent of very small firms, 27 percent of small firms 
and 11 percent of large firms do not follow a good manufacturing 
practice model. The survey results also show that 32 percent of 
vitamins and mineral establishments, 39 percent of amino acid/protein/
animal extract establishments, 41 percent of herbal and botanical 
establishments, and 59 percent of establishments not already 
classified, do not follow a good manufacturing practice model.
    Without industry-wide uniform requirements, some establishments may 
follow different practices but convey the message that they follow good 
manufacturing practices. In short, people who want to discriminate 
between establishments that use good practices and those that do not 
would not have sufficient information to do so. Another reason for our 
skepticism about universal voluntary adoption of good manufacturing 
practices is that good practices appear to be taken for granted by many 
consumers. Indeed, some consumers already believe that the Federal 
Government regulates the manufacturing practices of the industry, so 
firms lack an incentive to provide additional assurance (Ref. E3).
    Current or enhanced State and local regulations could bring about a 
reduction of potential harm from contaminated supplements. This 
alternative has the advantage that State and local governments can 
exercise more discretion when responding to local manufacturing 
conditions or consumer health practices than the Federal Government. 
Because most of the industry engages in interstate commerce, however, 
Federal regulations are appropriate. Also, Federal regulations would 
apply uniformly across the country, whereas State and local regulations 
might impose different standards on establishments that supply 
supplements across State and local boundaries.
    Litigation or the threat of litigation may help to bring about the 
goals of the proposed rule. The potential of costly litigation from the 
harm caused by deficient manufacturing practices creates an incentive 
for manufacturers to reduce the risks from defective products. However, 
we do not believe that litigation or the threat of litigation has 
created the incentives for all manufacturers to implement the 
manufacturing practices that we believe are necessary to avoid 
adulterated or misbranded products. As discussed earlier, not all 
surveyed dietary supplement manufacturers reported that they followed 
good manufacturing practices. Furthermore, in some cases it is 
difficult and costly to demonstrate to the courts that the harm to 
plaintiffs was

[[Page 12222]]

actually the result of poor manufacturing practices, making recourse to 
the courts sometimes impractical.
    In the absence of the proposed CGMP regulations, the burden of 
monitoring manufacturing practices would fall more heavily on 
consumers, despite the difficulties consumers face in monitoring 
manufacturers. Moreover, the proposed CGMP regulations are preventative 
and should ensure that problems are identified and dealt with during 
manufacturing, packaging, and holding, rather than after someone has 
consumed an unsafe product and experienced an adverse effect.
    b. Fewer requirements for vitamins and minerals. FDA could require 
more controls from establishments that manufacture, package, or hold 
plant or animal derived dietary ingredients such as amino acids, 
proteins, herbals, botanicals and other products not classified as 
vitamin and mineral manufacturers, packagers, or holders. The plant or 
animal derived dietary ingredients are probably characterized by 
greater variation in product quality than synthetically derived dietary 
ingredients. Under this option, the segment of the industry that 
manufacture, package, or hold products that are the most likely to have 
difficulty manufacturing or maintaining uniform product quality dietary 
ingredients would be required to follow the proposed testing and other 
production and process control requirements. Manufacturers of vitamins 
and minerals would be required to follow the sanitation, holding, and 
consumer complaint provisions only, they would not have to adopt 
manufacturing controls to ensure that products did not contain too much 
or too little of a vitamin or mineral.
    Plant or animal ingredients are likely to experience greater 
natural variation in product quality than synthetic compounds, so they 
may require the higher minimum standard of regulation contained in the 
proposed regulation. The advantage of this option is that fewer 
establishments will be affected as much; approximately 723 
establishments classified as manufacturers, packagers or holders of 
products other than vitamins and minerals, rather than the 1,566 
establishments estimated to be covered by the proposed regulation (see 
table 2 of this document). The compliance costs would therefore be 
lower. The disadvantage is that vitamin and mineral manufacturers also 
potentially manufacture products of variable quality, so the expected 
benefits from more consistent product quality would be reduced. 
Moreover, if dietary supplements contain too little of a vitamin or 
mineral consumers may not receive the intended health benefits, and if 
the dietary supplements contain too much of a vitamin or mineral they 
may experience illness or injury.
    We estimate that the benefits of this option would be approximately 
proportional to the ratio of recalled products that were classified as 
vitamins and minerals to all recalled dietary supplements products. 
Approximately 50 percent of the recalled products were vitamins and 
minerals so we estimate that this option would generate no more than 
$109 million in benefits. We assumed that the costs of this option 
would be proportional to the fraction of establishments that would be 
required to follow all of the proposed provisions and those that follow 
the reduced requirements with the total costs estimated for this 
proposal as shown in table 17 of this document. The estimated mean cost 
of the proposed regulation is $86 million (see table 19 of this 
document). The fraction of establishments required to follow all the 
provisions is .46 (= 723/1566). The fraction of establishments that 
would have reduced testing is .54 (= 843/1566). Testing is 
approximately 36 percent of the total costs. We estimate the total 
costs from this option to be $69 million ($86 million x .46 + $86 
million x .54 x (1-.36)).
    c. More restrictive CGMP regulations than the proposed regulations. 
One option is to propose (or finalize) more restrictive rules than the 
proposed CGMP regulations. Under this option, CGMP regulations could 
provide consumers with additional safeguards. Several of the largest 
manufacturers of dietary supplements now voluntarily comply with some 
of these additional safeguards (Ref. E2). The most significant 
additional provisions that would be required under this option are 
product quality testing for each incoming shipment lot of components 
and dietary ingredients, inprocess testing for contaminates at critical 
control points and mandatory written procedures for all of the various 
provisions of the proposed regulation.
    The advantage of this option is that the additional requirements 
provide safeguards that the essential safety and quality provisions are 
being followed. The disadvantage of this option is that it is more 
costly than the proposed rule, and we are not aware of any information 
that would show any additional verifiable health benefits.
    d. HACCP without the other elements of CGMP regulations. The agency 
could propose a requirement that manufacturers implement a HACCP (or 
HACCP like) system for the manufacturing of dietary supplements without 
the other elements of the proposed CGMP regulations. A critical control 
point is where production controls can be applied to reduce or 
eliminate hazards (including biological, chemical, or physical 
contamination) that may make dietary supplements unsafe.
    The advantage of an industry-wide HACCP program is that HACCP does 
not require manufacturers to follow detailed uniform requirements in 
order to achieve desirable outcomes. Manufacturers themselves determine 
for their specific products and processes how they will best eliminate, 
reduce, or control hazards in the manufacturing of dietary supplements.
    We have not designed a hypothetical HACCP system for the dietary 
supplement industry. For the purpose of generating estimates of costs 
and benefits, we assumed that a HACCP regulation for a dietary 
supplement manufacturer would be likely to encompass sanitation 
prerequisites that are met, writing a HACCP plan, and monitoring 
critical control points. The benefits and costs of the HACCP plan would 
be generated by controls for a narrower set of hazards in the 
manufacturing, packaging, and holding processes than those covered by 
this proposal, and would not include the other benefits and costs 
generated by the proposed rule especially the reduced consumer search 
costs, because uniform product quality would not necessarily be 
assured. The advantage of HACCP as an option to prevent product 
contamination is that it does not specify detailed manufacturing 
requirements. The disadvantage is that in the absence of uniform 
controls there would not be uniform minimum product quality across the 
industry and consumers would not derive the same benefits from lower 
search costs.
    e. Require final product testing only. FDA could propose that 
manufacturers test their finished products for identity, purity, 
quality, strength, and composition but not include any of the other 
mandatory provisions of the proposed regulation. The advantage of this 
option is that it would be the least costly option of those considered. 
Many firms already test some of their finished products, reducing the 
impact of this option. Approximately 69 percent of manufacturing plants 
conduct finished product testing and almost 65 percent of all finished 
batches in the industry are already tested using physical, chemical, 
microbiological, visual or organoleptic

[[Page 12223]]

testing techniques (Ref. E2). The problem with this option is that 
finished product testing alone cannot ensure product quality for some 
types of products. Not every finished product currently has a test that 
confirms identity, purity, quality, strength, or composition, 
especially for multiingredient products. Tests may not have been 
developed, or they may not be completely reliable, or they may not be 
capable of evaluating every type of product defect. Also, potentially 
lower cost alternatives to finished product testing--such as incoming 
component lot testing, inprocess testing, or both--might be available 
and desirable to firms as a means to protect the public. Moreover, 
finished product testing alone is not sufficient to prevent products 
with microbiological or chemical contamination from being discovered 
because it is possible that false negatives might occur, as when there 
is ``hotspot'' contamination within a batch. Preventative controls must 
be imposed to achieve that goal. Finally, finished product testing 
alone also will not facilitate trace backs when defective products are 
discovered in the marketplace, nor will it facilitate responsible 
investigations of consumer complaints. The estimated cost of this 
option is lower than that of the other options, but it does not 
generate the full range of benefits provided by the proposed rule.
    f. Regulate only high-risk products. FDA could propose CGMP 
regulations that would cover only high-risk products. The advantage of 
this option is that it would impose lower costs than the proposed rule, 
but (if all risky products could be identified and regulated) generate 
the same level of benefits. Only those establishments that manufacture 
high-risk products or have high-risk hazards would incur the costs of 
adopting CGMP regulations. High-risk might be defined as those products 
most likely to be contaminated, or suffer other product defects. There 
are two problems with this option. Adverse event reporting is not 
mandatory, so significant underreporting is expected. Also, it is 
possible that the confirmed illnesses and other problems linked to 
particular dietary supplements may be those most easily traced, rather 
than those with the highest risk. High levels of identified problems 
may not be closely correlated with high levels of risk. In other words, 
problems associated with the known defective products may or may not be 
correlated with the highest risk. Without more data and risk 
assessments, it would be difficult to distinguish what risks may be 
associated with particular dietary supplements. We therefore have no 
basis upon which to begin a full evaluation of what the high-risk 
products are or may be.
3. Coverage of the Proposed Rule
    The proposed rule would cover establishments that manufacture, 
package, hold dietary ingredients or dietary supplements. Tables 2, 3, 
and 4 of this document list the estimated number of covered 
manufacturers, packagers, dietary ingredient suppliers, holders, and 
other establishments. Table 2 of this document shows the number of 
covered establishments by product type and size. A small business, 
based on the Small Business Administration definition, is any firm with 
500 or fewer employees. For purposes of analysis, we defined very small 
establishments as having fewer than 20 employees. Table 3 of this 
document shows the number of establishments categorized as 
manufacturers, ingredient suppliers, repackers or relabelers, holders 
whose primary business is dietary ingredients or dietary supplements, 
and other (although not including other holders and distributors). 
Table 4 of this document shows our estimate of the number of general 
warehouses and wholesalers that hold dietary supplements.

               Table 2.--Covered Establishments by Product Type and Size From Dietary Supplement Enhanced Establishment Database (DS-EED)
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                  Very
                         Product type                            small       %       Small       %       Large       %       Unknown      %       Total
--------------------------------------------------------------------------------------------------------------------------------------------------------
Vitamins and Minerals.........................................      252       29.8      223       26.5       78        9.2       290       34.5      843
Amino Acids, Proteins.........................................       21       31.0       16       23.0        6        6.9        27       39.1       69
Herbals and botanicals........................................      148       42.6       46       13.2        5        1.1       150       43.1      348
Supplements not already classified............................       93       30.4       66       21.6       20        6.5       127       41.6      306
                                                               ----------
    Total.....................................................      514       32.8      351       22.4      106        6.8       594       38.0    1,566
--------------------------------------------------------------------------------------------------------------------------------------------------------


    Table 3.--Covered Establishments by Type of Operation From DS-EED
------------------------------------------------------------------------
                                             Number of      Percent of
           Establishment type             establishments  establishments
------------------------------------------------------------------------
Manufacturer............................          1,228            78.4
Dietary ingredient supplier.............            106             6.7
Repacker; relabeler.....................             26             1.7
Holder..................................            114             7.3
Establishments not already classified...             92             5.9
                                         -----------------
    Total...............................          1,566           100.0
------------------------------------------------------------------------


     Table 4.--Covered Establishments That Hold Dietary Supplements
------------------------------------------------------------------------
                                                             Number of
        Type of holders            Source and SIC code    establishments
------------------------------------------------------------------------
Grocery Wholesalers or Drug      Dunn and Bradstreet:            25,527
 Wholesalers.                     5122, 5141.
Food or Drug Warehouse.........  Dunn and Bradstreet: N/            738
                                  A.
Miscellaneous Food or Drug       Dunn and Bradstreet:               238
 Warehouse.                       4225, 4226, 5912,
                                  5499, 5411, 5122,
                                  5141, 5149, 5399,
                                  5311, and 5331.

[[Page 12224]]

Dietary Supplement.............  DS-EED.................            114
                                --------------------------
    Total......................  .......................         26,617
------------------------------------------------------------------------

    We consulted several sources to estimate the number of 
establishments reported in tables 2, 3, and 4 of this document. The 
number shown in tables 2 and 3 of this document, 1,566, is the 
estimated number of establishments in the DS-EED that manufacture, 
repackage, supply dietary ingredients, or hold dietary supplement 
products in the United States. RTI developed the DS-EED using FDA's 
Official Establishment Inventory (OEI) and supplemented that source 
with information from trade organizations, trade shows, and electronic 
databases (Refs. E1 and E2).
    The number of establishments in the DS-EED that hold dietary 
supplements is not the total number of holders covered by the proposed 
regulation. The holding establishments in the DS-EED identified holding 
dietary supplements as their primary business. To estimate the total 
number of establishments that could hold dietary ingredients or dietary 
supplements but do not consider dietary supplements as their primary 
business, we performed three searches of firms that are listed with Dun 
and Bradstreet's Dialog database. We first looked for a count of firms 
that had standard industrial classification (SIC) codes for wholesalers 
of groceries or drugs. Next we looked for a count of firms that met the 
description of warehouses of groceries or drugs (no SIC codes were 
used). Finally, we looked for a count of any firms that had both 
warehouse SIC codes and miscellaneous drug stores, food stores, 
sundries, and general merchandise (SIC 4225, 4226, 5912, 5499, 5411, 
5122, 5141, 5149, 5399, 5311, and 5331). The results are shown in table 
4 of this document. We concluded that the total number of 
establishments in this category that could hold dietary ingredients or 
dietary supplements and would be covered by the regulation was 
approximately the sum of the numbers counted in the three searches, or 
26,617.
    The number of establishments that hold dietary ingredients or 
dietary supplements includes retailers that sell dietary supplements to 
consumers, and transporters of dietary ingredients and dietary 
supplements. We made no effort to determine the number of such holders, 
because the proposed requirements do not apply to retailers and 
transporters. We believe that retailers and transporters may 
voluntarily adopt provisions related to the holding of these products 
and thus there may be changes in the marketplace with accompanying 
costs and benefits. However, we expect that the only retailers and 
transporters that will voluntarily adopt the proposed requirements are 
those that expect the private benefits of adoption will exceed the 
private costs.
4. Baseline Practices
    a. Consumer baseline practices. Baseline consumer and manufacturer 
practices, governed by current market forces and existing government 
regulations, give rise to the current risks associated with the 
manufacturing of dietary supplements. When determining baseline 
manufacturing practices, it is necessary to estimate both the practices 
that are used now, as well as the likely changes in manufacturing 
practices that will occur even in the absence of new regulations. The 
risks to consumers from these products can be associated with a 
combination of consumption habits, the contamination of the products, 
or both. Contamination may be caused by current manufacturing 
practices. Consumption is influenced by the price and quality of 
dietary supplements, set by the interaction of market participants. 
Finally, changes in practices of either consumers or manufacturers 
caused by new regulatory requirements will give rise to changes in 
risks, as estimated by changes in costs and benefits.
    The consumption of dietary supplements has grown in recent years. 
Consumers report that they are using a wider range of product types, 
and that they are using dietary supplements for more reasons than they 
were in the past.
    Table 5 of this document illustrates the rapid sales growth of the 
dietary supplement industry from 1994 to 2000. Panel A of table 5 of 
this document shows annual sales of three general categories of dietary 
supplements, a measure of the market size of the supplement industry. 
Annual increases in sales of herbals and botanicals were the greatest, 
averaging 18 percent per year, while annual increases in sales of 
supplements that were neither vitamins and minerals nor herbals and 
botanicals increased less, averaging 11 percent per year. The lowest 
annual sales increases were for vitamins and minerals, averaging 8 
percent per year. For all dietary supplements combined, sales increased 
an average of 12 percent a year since 1994 (not shown on the table).
    While the sales growth shown in table 5 of this document, Panel A, 
is impressive, only part of this apparent growth represents increased 
use. Population growth and rising prices also contributed to the 
apparent growth. The real (growth inflation-adjusted) increase in 
dietary supplement prices is estimated by subtracting the inflation 
rate from the rate of price increases of dietary supplements (Ref. E4). 
As shown in table 5 of this document, Panel B, between 1995 and 1997 
the real price of vitamins and minerals and supplements other than 
vitamins and minerals all increased. Rising real price indicates that 
demand is growing rapidly.
    Table 5 of this document, Panel C, shows estimated annual increases 
in per capita consumption of dietary supplements.\1\ As shown in table 
5 of this document, Panel C, the estimated per capita consumption of 
the different categories of dietary supplements has increased since 
1994.
---------------------------------------------------------------------------

    \1\ An index measuring per capita consumption of dietary 
supplements can be derived using the following equation: 
PCCt = [1,000 x Salest]/[POP x Pt 
], where, t = year index; PCCt = per capita consumption 
(# of unit sales); Sales = millions of dollars of sales; 
POPt = thousands of U.S. residents; Pt = 
average price of supplement. In the formula, we measure consumption 
as the number of dietary supplement units (bottles, packages, etc.) 
sold per U.S. resident for a given year.
---------------------------------------------------------------------------

    For the consumption estimates in table 5 of this document, we 
averaged dietary supplement use over the entire U.S. population, 275 
million. In table 6 of this document, we included estimated average 
supplement use for the population of supplement users, 160 million 
(Ref. E13). The three panels in table 6 of this document show the 
annual consumption per supplement user and the annual change in 
consumption per supplement user for

[[Page 12225]]

vitamins and minerals, herbals and botanicals, and supplements other 
than vitamins and minerals and herbals and botanicals. Table 6 of this 
document also shows that during this period the proportion of consumers 
using supplements increased faster than the average consumption for the 
total population. The surprising implication of this result is that 
consumption per user has apparently declined since 1994.
    One limitation of the estimates in table 6 of this document is that 
prevalence of supplement use is based on the proportion of U.S. adults 
consuming supplements, while the per capita consumption figures are 
based on the entire U.S. population. Nonetheless, we do not have any 
reason to believe that the estimated trend in consumption per user is 
biased. This trend, expressed as the percentage change in consumption 
per user, is negative for all segments of the dietary supplement 
industry since 1994. The large and rising number of consumers accounts 
for the growing size of the dietary supplement industry.

                              Table 5.--Growth in Market Size and per Capita Consumption of Dietary Supplements, 1994-2000
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                   1994         1995         1996         1997         1998         1999         2000
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                  Panel A--Nominal Market (Millions of Current Dollars)
--------------------------------------------------------------------------------------------------------------------------------------------------------
Vitamins.....................................................        3,960        4,220        4,780        5,190        5,550        5,940        6,360
Growth rate (percent)........................................  ...........         6.57        13.27         8.58         6.94         7.03         7.07
Minerals.....................................................          700          800          900        1,070        1,160        1,250        1,350
Growth rate (percent)........................................  ...........         14.0         13.0         19.0          8.0          8.0          8.0
Herbals and Botanicals.......................................        2,070        2,530        2,990        3,530        4,170        4,840        5,520
Growth rate (percent)........................................  ...........        22.22        18.18        18.06        18.13        16.07        14.05
Supplements other than vitamins/minerals and botanicals......        2,070        2,290        2,620        2,890        3,180        3,490        3,840
Growth rate (percent)........................................  ...........        10.63        14.41        10.31        10.03         9.75        10.03
    Total....................................................        8,080        9,840       11,290       12,680       14,060       15,520       17,070
Growth rate (percent)........................................  ...........         12.0         15.0         12.0         11.0         10.0         10.0
--------------------------------------------------------------
                                                                     Panel B--Prices
--------------------------------------------------------------------------------------------------------------------------------------------------------
Consumer price index-units (percent).........................        148.5        152.5        157.0        160.5        163.2        166.7  ...........
Inflation rate (percent).....................................         2.56         2.76        2.957         2.23         1.68         2.14         2.39
Vitamins and minerals........................................  ...........  ...........  ...........  ...........  ...........  ...........  ...........
Average nominal price (IRI)..................................        $6.20        $6.50        $6.87        $7.34        $7.54        $7.78        $8.05
Nominal price increase (percent).............................         2.69         4.84         5.69         6.84         2.72         3.18         3.43
Real price increase (percent)................................         5.25         2.08         2.74         4.61         1.04         1.04         1.04
Supplements other than vitamins and minerals:
    Average nominal price....................................        $6.20        $6.50        $6.87        $7.34        $7.70        $8.11        $8.56
    Nominal price increase (percent).........................         5.80         4.84         5.69         6.84         4.85         5.31         5.56
    Real price increase (percent)............................         3.24         2.08         2.74         4.61         3.17         3.17         3.17
--------------------------------------------------------------
                                        Panel C--Per Capita Consumption (Number of Units Sold Per U.S. Resident)
--------------------------------------------------------------------------------------------------------------------------------------------------------
Vitamin/mineral sales........................................         2.45         2.47         2.62         2.64         2.72         2.80         2.87
Growth (percent).............................................  ...........         0.69         6.19         0.66         3.12         2.74         2.55
Herbals sales................................................         1.28         1.48         1.64         1.80         2.00         2.19         2.34
Growth (percent).............................................  ...........        15.48        10.79         9.45        11.60         9.17         7.03
Supplements other than vitamins and minerals and herbals              1.28         1.34         1.44         1.47         1.53         1.58         1.63
 sales.......................................................
Growth (percent).............................................  ...........         4.53         7.26         2.26         3.95         3.23         3.25
--------------------------------------------------------------------------------------------------------------------------------------------------------

  Table 6.--Comparison of Consumption per Person With Consumption per User: Evidence That the Dietary Supplement Market Is Becoming Broader Not Deeper
--------------------------------------------------------------------------------------------------------------------------------------------------------
                        Average Growth                             1994         1995         1996         1997         1998         1999      1994-2000
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                A. Vitamins and Minerals
--------------------------------------------------------------------------------------------------------------------------------------------------------
Per capita consumption (units per U.S. resident).............         2.45         2.47         2.62         2.64         2.72         2.80  ...........
% Growth.....................................................  ...........         0.69         6.19         0.66         3.12        2.74%        2.68%
Consumption prevalence (percent).............................  ...........        47.70         54.0         61.0         70.0         79.0  ...........
Reference....................................................  ...........      Ref. E6      Ref. E6      Ref. E6      Ref. E6      Ref. E7  ...........
% Growth.....................................................  ...........  ...........        13.44        13.44        13.44        13.44        13.44
Consumption per user (units).................................  ...........         5.18         4.85         4.30         3.91         3.54  ...........
% Growth.....................................................  ...........  ...........        -6.39       -11.27        -9.10        -9.43        -9.05
--------------------------------------------------------------
                                                                   1994         1995         1996         1997         1998         1999      1994-1999
--------------------------------------------------------------
                                                                B. Herbals and Botanicals
--------------------------------------------------------------------------------------------------------------------------------------------------------
Per capita consumption (units per U.S. resident).............         1.28         1.48         1.64         1.80         2.00         2.19  ...........

[[Page 12226]]

% Growth.....................................................  ...........        15.48        10.79         9.45        11.60         9.17        11.30
Consumption prevalence (percent).............................         8.20        12.10        12.10        12.10           28           49  ...........
Reference....................................................      Ref. E8      Ref. E8      Ref. E8      Ref. E9     Ref. E10      Ref. E7  ...........
% Growth.....................................................  ...........        47.56         0.00         0.00       131.40        75.00        50.79
Consumption per user (units).................................        15.64        12.24        13.56        14.84         7.16         4.47  ...........
% Growth.....................................................  ...........       -21.74        10.79         9.45       -51.77       -37.62      -18.18%
--------------------------------------------------------------
                                       C. Supplements Other than Vitamins and Minerals and Herbals and Botanicals
--------------------------------------------------------------------------------------------------------------------------------------------------------
Per capita consumption (units per U.S. resident).............         1.28         1.34         1.44         1.47         1.53         1.58  ...........
% Growth.....................................................  ...........         4.53         7.26         2.26         3.95         3.23         4.24
Consumption prevalence (percent).............................          5.1          8.8         11.2         14.2         18.1         23.0  ...........
Reference....................................................      Ref. E8      Ref. E8      Ref. E8      Ref. E8      Ref. E8      Ref. E7  ...........
% Growth.....................................................  ...........        72.55        27.15        27.15        27.15        27.14        36.23
Consumption per user (units).................................        25.15        15.24        12.85        10.34         8.45         6.86  ...........
% Growth.....................................................  ...........       -39.42       -15.64       -19.58       -18.25       -18.81       -22.34
--------------------------------------------------------------------------------------------------------------------------------------------------------

    b. Manufacturer's baseline practices. FDA contracted with RTI to 
conduct a survey of the dietary supplement industry to learn about both 
baseline (existing) manufacturing practices and the existing standards 
used for manufacturing dietary ingredients and dietary supplements 
(Ref. E2). A sample of 966 dietary supplement establishments from the 
DS-EED database was selected from an estimated eligible population of 
1,566 firms in the industry. The sample was stratified by 
manufacturer's product type and the size of firm in the industry. 
Stratification helps ensure that estimates of the subpopulations are 
more precise. Establishments that were stratified by manufacturer's 
product type were classified as primarily: (1) Vitamins and minerals; 
(2) amino acids, proteins, or animal extracts; (3) herbals and 
botanicals; or (4) all other product types not already classified. The 
product type strata were further stratified by four size categories: 
(1) Very small, (2) small, (3) large, and (4) unknown. This 
categorization generated 16 sampling strata.
    The contractor, RTI, sent each of the 966 firms in the sample a 
lead letter on FDA letterhead and a one-page brochure to explain the 
purpose of the survey, the value of the establishment's participation, 
and the agency's confidentiality procedures. Following the mailing, RTI 
placed telephone calls to each establishment to screen for eligibility 
and to recruit eligible establishments for the mail survey. To be 
eligible for the survey, establishments had to currently manufacture, 
repackage, supply dietary ingredients, hold, import or export dietary 
supplements for human consumption. Almost 50 percent of the 
establishments sampled were not eligible for the survey because they 
were no longer in operation at the listed address or did not handle any 
dietary supplements or ingredients for human consumption.
    To achieve the highest possible response rate, RTI operated a toll-
free help line and attempted to contact each establishment up to eight 
times before assigning a disposition of nonresponse. RTI also attempted 
up to two refusals conversions, which are attempts to persuade firms 
that declined to answer the survey to respond. The survey was conducted 
over a 10-week period, November 29, 1999, to February 4, 2000. There 
were a total of 238 completed surveys, resulting in a final disposition 
of: (1) An overall eligibility rate of close to 50 percent, and (2) a 
response rate of 50 percent.
    Determining baseline practices is necessary in order to determine 
the new activities that are likely to take place as a result of 
implementation of this proposed rule. Each of the new activities 
potentially brought about by the proposed rule has both a marginal (or 
incremental) cost and a marginal (or incremental) benefit. These 
incremental costs and benefits of likely new activities form the basis 
of our economic analysis of the proposed rule.
    The survey asked establishments a series of questions about 
existing practices; we used the responses to estimate how many 
establishments in the industry already operated in accordance with the 
requirements of the proposed regulation. One key assumption in this 
analysis is that no firms are expected to stop CGMPs and no firms are 
expected to start good manufacturing practices in the absence of this 
rule. The universe for the survey includes the establishments discussed 
in section VII.B.3 of this document. If firms start good manufacturing 
practices in the absence of this rule, both the costs and benefits of 
the rule would be less than we estimate. If firms were to stop in the 
absence of the rule, both the costs and benefits would be more than we 
estimate. We lack information about the trend in the industry, so we 
assumed that the survey reflects both the current and future practices 
in the industry. We request comment or information about the industry 
trend in adopting good manufacturing practices.
    i. Stratification. The survey was stratified by product type and 
establishment size. Stratification ensures that samples are 
representative of the industry population.\2\ The subdivisions of the 
population of interest here were establishment size (by the number of 
employees) and product type, because these characteristics are likely 
to influence whether an establishment already has adopted the practices 
that would be required by the regulation. The DS-EED includes nine 
product types: (1) Vitamins and minerals; (2) herbals and botanicals; 
(3) herbal and botanical extracts; (4) amino acids; (5) proteins; (6) 
animal extracts; (7) tea like products; (8) concentrates, metabolites, 
or constituents; and (9) supplements not already classified (all other 
supplements). Establishments may produce more than one product type; 
establishments with multiple product types were, however, only 
classified in one category. For stratification and reporting purposes, 
we defined the

[[Page 12227]]

following four mutually exclusive categories of dietary supplements:
---------------------------------------------------------------------------

    \2\ Stratification is a subdivision of the population of 
establishments in the dietary supplement industry by a unique 
characteristic such as product type or number of employees.
---------------------------------------------------------------------------

    1. Vitamins and minerals (includes establishments that may also 
manufacture, package, or hold herbals and botanicals, amino acids, 
proteins, or animal extracts but predominately manufacture vitamins and 
minerals);
    2. Amino acids, proteins, and animal extracts (includes 
establishments that also manufacture, package or hold herbals and 
botanicals, including extracts; excludes establishments already 
classified as vitamins and minerals);
    3. Herbals and botanicals, including extracts (excludes 
establishments already classified as ``vitamins and minerals'' or 
``amino acids, proteins, or animal extracts''); and
    4. Supplements not already classified (all other product types).
    We further stratified each of the four product categories into four 
size categories, very small, small, large, and unknown--resulting in 16 
sampling strata. We classified each establishment into one mutually 
exclusive industry category (manufacturer, dietary ingredient supplier, 
repacker/relabeler, holder, or establishment not already classified). 
Establishments that manufacture supplements and also supply, repack, or 
hold dietary supplements or ingredients were classified as 
manufacturers.
    ii. Size stratification. The Small Business Administration 
classifies companies as ``small'' based on the size of the entire 
company, including both parent and subsidiaries. If firms that 
manufacture dietary supplements have 500 or fewer employees, they are 
classified as small. Because the DS-EED data on size are only for 
specific establishments and not parent firms, we had to obtain parent 
company information on employment or revenue to correctly classify each 
establishment as part of a small or large company. To obtain parent 
company data for establishments in the survey universe, we sent InfoUSA 
\3\ the DS-EED data records (N = 2,004) and requested the name, 
address, primary SIC, employment size (in ranges), and revenue (in 
ranges) of parent company firms with establishments in the survey 
universe. InfoUSA matched 1,219 of the 2,004 records in the DS-EED to 
their U.S. database of 10.3 million businesses. Of the 1,219 matched 
records, 31 records were found to be duplicates of another record and 
were removed, leaving 1,188 matched records and 1,566 total records in 
the sampling frame. The nonmatched records did not match because: (1) 
They were recently established businesses, (2) they were out of 
business, or (3) they had recently changed their names or addresses. 
Because data on revenue or employment size were not available for the 
nonmatched records, we created an ``unknown'' stratum for these 
establishments. The survey of practices collected information on 
employment that allowed us to classify some of these establishments by 
size for the analysis.
---------------------------------------------------------------------------

    \3\ InfoUSA is a publicly held company that creates proprietary 
business databases. Their database includes such information as: 
Company name, address, phone number, fax number, estimated sales, 
volume, number of employees, type of business (SIC code or yellow 
page heading), key contact names, and titles.
---------------------------------------------------------------------------

    Of the 1,188 matched records, 180 were linked to parents. The 
parent company data for these 180 establishments were merged with the 
survey universe. The remaining 1,008 records did not link to an 
ultimate parent company. For these records, the establishment and 
parent company were the same entity, so we used establishment level 
data to classify size. We classified each of the establishments in the 
survey universe as part of very small, small, or large businesses based 
on the employment size or annual revenues of each establishment's 
parent company. If an establishment or its parent company had 500 or 
fewer employees or sales less than $20 million (if data on employment 
were not available), then the establishment was classified as small. An 
establishment was classified as very small if the number of employees 
was less than 20.
    iii. Survey response. Table 7 of this document presents the number 
of establishments surveyed, stratified by the four product types and by 
size. Although the sample allocation was designed to yield 400 
completed surveys, we received only 238 completed mail surveys. The 
number of respondents was fewer than expected because the number of 
establishments that were ineligible was greater than we expected and 
because some establishments did not respond to the survey after 
agreeing to participate. Ineligible establishments are those that no 
longer produce dietary supplements because they have gone out of 
business or changed product lines, or they have moved and could not be 
located. Despite receiving fewer responses than planned, the confidence 
level for the final results allowed us to make meaningful inferences 
regarding the industry. For example, 65 percent of the establishments 
surveyed responded that they followed published good manufacturing 
practice models; the 95 percent confidence interval was 56 to 72 
percent. By size category, 52 percent of very small, 73 percent of 
small, and 89 percent of large establishments responded that they 
followed published good manufacturing practice models (Ref. E2). 
Although we do not suggest that these percentages are precise, they do 
tell a plausible story of the current use of good manufacturing 
practice models in the supplement industry: The use of good 
manufacturing practice models appears to be widespread but far from 
universal, with use more likely the larger the establishment.

                            Table 7.--Number of Completed Surveys by Sampling Strata
----------------------------------------------------------------------------------------------------------------
                                                                                      Size
                                                               -------------------------------------------------
                         Product type                             Very
                                                                  small     Small     Large    Unknown    Total
----------------------------------------------------------------------------------------------------------------
Vitamins and minerals.........................................        19        39        13         1        72
Amino acids, proteins.........................................         8         7         0         5        20
Herbals and botanicals, including extracts....................        58        25         0        30       113
Supplements not already classified............................        14        13         2         4        33
                                                               -----------
    Total.....................................................        99        84        15        40       238
----------------------------------------------------------------------------------------------------------------

    The mean survey results reflect the degree of uncertainty 
associated with each practice. The use of a survey for this economic 
analysis often required the use of the survey answers from more than 
one question to assess the impact

[[Page 12228]]

of each proposed provision. For example, answers to questions about 
testing herbals might have been combined with questions about whether 
the firms manufactured herbals. Some highlights of the survey are:
    ? Plant characteristics: Manufacturers account for 62 percent 
of the total firms and 36 percent of manufacturers produce vitamins and 
minerals as their primary product.
    ? Use of published good manufacturing practice model: 65 
percent of all firms follow some type of good manufacturing practice 
model, primarily food good manufacturing practices; 28 percent follow 
the NNFA good manufacturing practices and 31 percent follow FDA's drug 
good manufacturing practice requirements.
    ? Personnel: 67 percent of all establishments maintain 
records of personnel education, training, or experience.
    ? Quality control: 85 percent of all establishments have a 
unit or person responsible for quality control. Almost 80 percent of 
all manufacturers conduct at least some type of identity tests on 
incoming components and dietary ingredients and 96 percent of these 
firms also conduct some type of contamination test; 63 percent conduct 
some type of potency test. Nearly 70 percent conduct tests on inprocess 
materials or finished products. Of these firms, 97 percent conduct 
identity tests, 94 percent conduct contamination tests and 72 percent 
conduct potency tests. Asked whether firms hold reserve samples of each 
finished batch, 75 percent answered yes. Of the plants that have 
production processes, 70 percent use production and process controls 
that identify the points, steps, or stages in the manufacturing process 
to prevent adulteration. Almost 68 percent of all incoming ingredient 
or component lots are tested now and almost 70 percent of inprocess or 
finished product batches are tested in some manner.
    ? Warehousing: 70 percent of warehouses have temperature 
controls and 22 percent have humidity controls.
    ? Consumer complaints: Only 19 percent report incidents to 
FDA.
5. Baseline Risk
    The current number of illnesses caused by poor manufacturing 
practices requires data linking illnesses directly to poor practices. 
Without direct evidence on the number of illnesses caused by poor 
manufacturing practices, we had to use an indirect approach. There are 
two indirect ways to estimate the number of illnesses caused by 
defective products:
    ? We could take the number of reported cases and multiply by 
a factor to account for underreporting.
    ? We could take the number of defective products and multiply 
by the probability of illness for the given defect.
    In an ideal analysis, we would estimate the baseline both ways and 
then compare them. For the analysis of illnesses from poor 
manufacturing practices, however, we did not have sufficient data to 
perform either type of baseline estimate.
    We looked at many sources for information, including medical and 
other literature on adverse events, information from poison control 
centers, reports to the agency, popular newspaper and magazine 
articles, and surveys of users. The literature review was conducted 
using Medline, Healthstar, Aidsline, Cancerlit, and OldMedline (Ref. 
E12). We found evidence of many adverse events associated with dietary 
supplements. For example, one recent survey found that 12 percent of 
consumers (about 11.9 million) who have used an herbal remedy claim to 
have suffered from side effects or other adverse reactions (Ref. E13). 
The American Association of Poison Control Centers received 6,914 
reports on dietary supplements in 1998 (Ref. E14). In a recent survey, 
46 percent of respondents answered that people get sick from dietary 
supplements ``often'' or ``sometimes'' (Ref. E3). In addition, the 
agency has received many voluntary reports of illnesses caused by 
dietary supplements (Ref. E15). The vast majority of the illnesses 
described in the sources we consulted, however, are reported as 
associated with the ingredients used in the products themselves, not 
with poor manufacturing processes. We have no direct evidence on what 
fraction of illnesses can be attributed to manufacturing processes. The 
anecdotal evidence implies that many illnesses could have been caused 
by poor manufacturing processes, but with a few exceptions, no evidence 
explicitly links illnesses to these manufacturing processes.
    The agency's recall records are more useful than the reports on 
illnesses, because the class 1 and class 2 recalls all involve 
defective products that could have caused illness if ingested. The 
major public health events that have been linked to poor manufacturing 
processes show up in the list of dietary supplements recalled. Although 
the recall data cannot be linked directly to illness data, we have 
found anecdotes, surveys, and some medical literature on illnesses that 
could be caused by avoidable manufacturing mistakes. We have recall 
data that show that manufacturing mistakes exist, so we can construct a 
possible link between manufacturing mistakes and potential illnesses or 
injuries. The number of illnesses associated with a recall is both 
variable and uncertain, and could be anything from zero to quite large. 
We concluded that one illness would not be an implausibly high average 
for a recall, so we assumed that a recalled product could be a proxy 
for a single reported illness associated with a defective product. We 
ask for comments on this assumption.
    Because there are no well established systems for the notification 
of adverse health events related to dietary supplements, and some 
significant barriers to reporting, we assume that unreported illnesses 
caused by poor manufacturing practices are substantially greater than 
reported illnesses. We relied on Ref. E16 to estimate a more precise 
relationship between reported and unreported rates. Based on empirical 
data for drug and vaccine reporting rates among other studies, the 
author of Ref. E16 determined that for dietary supplements, reported 
illnesses represent at best approximately 1 percent of total illnesses 
(Ref. E16). A similar multiplier of 100 linking known cases of 
foodborne illness to total incidence is often used. We assume that 
reporting adverse health events due to poorly manufactured dietary 
supplements would occur at the same proportion as adverse health events 
caused for other reasons by dietary supplements. We show the 
sensitivity of benefits to the choice of multiplier below, in the 
uncertainty and sensitivity analysis of our results.
    The outbreak of eosinophilia-myalgia syndrome (EMS) resulting from 
contaminated L-Tryptophan resulted in the recall of the contaminated 
products. In part based on this example, we assume that product recalls 
can indicate when there are adverse health events. We also assume that 
the reported class 1 and class 2 recalls that have occurred over the 
last 10 years represent the number and type of recalls that will occur 
in the future but for the implementation of this regulation. From 1990 
through 1999, the agency received reports on an annual average of 13 
class 1 and class 2 recalls of dietary supplements. If each recall is a 
proxy for a reported illness, then the total number of unreported 
illnesses per year is approximately 1,300. Obviously, to the extent 
that products are successfully recalled, illnesses will be avoided. Our 
assumption is that the recall occurs

[[Page 12229]]

because at best one person on average has been made ill. We recognize 
that our procedure generated highly uncertain estimates of the number 
of illnesses. The use of recalls to estimate reported and unreported 
illnesses probably generated a distribution of illnesses below the 
``true'' distribution, because many illnesses occur that are not linked 
to recalls and are never reported. We were not able to determine even 
the approximate size of the underestimation from this procedure.
    We estimated the monetary value of the health benefits from CGMP 
regulations by multiplying the number of illnesses prevented by the 
health costs associated with an illness. The health benefits associated 
with preventing an illness come from: (1) Preventing the loss of 
productivity, (2) the reduction in pain and suffering, and (3) the 
reduction in expenditures on medical treatment. We measured lost 
productivity indirectly with measures of functional state, which 
includes measures of physical function. We estimated the losses caused 
by pain and suffering with a symptom-problem index. We used direct 
measures of medical costs, such as payments to physicians and 
hospitals.\4\
---------------------------------------------------------------------------

    \4\ The cost of a hospital day is from the Health Care Financing 
Agency's Indicator Tables. It is the amount per patient day in 1997, 
adjusted to 1999 dollars. See Ref. E17.
---------------------------------------------------------------------------

    Table 8 of this document contains summaries of our measures of the 
health effects potentially caused by known instances of defective 
products associated with poor manufacturing processes. We estimated the 
health loss per day for the different levels of illness severity by 
summing the lost productivity (as measured by functional state) and the 
loss from pain and suffering (as measured by the symptom-problem index 
\5\). These losses per day can be interpreted as the difference between 
a day of normal health, where normal is defined as the population's 
health not affected by these products, and a day of suffering from the 
health conditions caused by these defective products. The numerical 
scale is a relative baseline that rests on the notion of a quality-
adjusted life day (QALD). The QALD for a day of normal health equals 1; 
the QALD for death equals 0. The loss of QALDs per illness equals the 
daily loss multiplied by the number of days the illness lasts. We 
converted QALDs to dollars by multiplying the index numbers by the 
value of a statistical life day and adding the direct medical costs.
---------------------------------------------------------------------------

    \5\ Functional Status Code is a measure of lost mobility (MOB), 
physical activity (PAC) and social activity (SOC). Lost MOB might 
mean an inability to drive a car. Lost PAC might mean walking with 
physical limitations. Lost SOC might mean self-care is not possible. 
Symptom-problem health utility index is a weighted measure of the 
cost of each symptom. For example, a sick or upset stomach has a 
utility weight of .290.

                      Table 8.--Summary of Health Effects Based on Potential Illness Associated With Recalls Between 1990 and 1999
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                   Number                              Frequency of     Quality    Duration of    Medical    Health cost
                Problem                 Class of     of              Outcomes             illness      adjusted      illness      cost ($)     ($) per
                                         recall    recalls                               (percent)     life day       (days)     per event      event
--------------------------------------------------------------------------------------------------------------------------------------------------------
Hypervitaminosis A....................         1         2  .........................        100            0.472            3           84          936
Salmonella............................         1         4  Mild.....................         93.8          0.473            2            0          534
                                        ........  ........  Moderate.................          5            0.473            5          800        2,223
                                        ........  ........  Severe...................          1.2          0.563           17        9,100       14,859
                                        ........  ........  Reactive arthritis (short          2            0.42            25          100        6,438
                                                             term).
                                        ........  ........  Reactive arthritis (long           1            0.42         5,223          400    1,320,252
                                                             term).
                                               2         4  Death....................          0.04  ............  ...........        9,100    5,009,100
Klebsiella pneumonia..................         1         1  Severe...................         85     ............  ...........        6,235       10,650
                                        ........  ........  Death....................         15     ............  ...........        6,235    5,006,325
Selenium poisoning....................         1         1  Low doses................         50            0.482            3           84          954
                                        ........  ........  Severe...................         35            0.482            3        2,578        4,448
                                        ........  ........  Death....................         15     ............  ...........        2,578    5,002,578
Stannous fluoride.....................         1         1  Acute....................        100            0.473            3           84          938
                                               2         1  .........................  ............         0.473            3           84          938
Eosinophilia-myalgia syndrome.........         1         7  Mild.....................         47            0.482        5,223        1,176    1,515,863
                                        ........  ........  Moderate.................         50            0.482           60           84       17,484
                                               2        41  Severe...................         10     ............  ...........       14,964       27,394
Glass fragments.......................         2         1  Dental injury, simple....         50            0.231            1          139  ...........
                                        ........  ........  Dental injury,                    12     ............  ...........        3,741  ...........
                                                             complicated.
                                        ........  ........  Oral emergency...........         12     ............  ...........        3,741        6,428
                                        ........  ........  Tracheo-esophageal                25     ............  ...........  ...........          290
                                                             obstruction.
                                        ........  ........  Esophageal performation..          1     ............  ...........       14,964       23,343
Hypervitaminosis D....................         2         1  .........................        100            0.473            3          168        1,022
Pyridoxine (vitamin B6)...............         2         2  .........................        100            0.482           30          168        8,868
Super-potent zinc.....................         2         1  Mild.....................         50     ............  ...........  ...........          285
                                        ........  ........  Moderate.................         40     ............  ...........  ...........          596
                                        ........  ........  Severe...................         10     ............  ...........        1,247        3,347
Niacin................................         2         1  .........................        100     ............  ...........           84        4,258
Yellow #5 (undeclared)........         2         5  Mild allergic reaction...         90            0.44             2            0          529
                                        ........  ........  Severe allergic reaction.         10     ............  ...........        2,494        3,346

[[Page 12230]]

                                        ........  ........  Contact dermatitis.......         50     ............  ...........           84        1,205
Yellow #6, red #40,            2         1  Abdominal cramps.........         10            0.473            3           84          938
 blue #2 (undeclared).
                                        ........  ........  Contact dermatitis.......         90     ............  ...........           84        1,205
Copper salts..........................         2         1  .........................        100            0.473            1           84          369
Digitalis.............................         1        33  Mild.....................         94.9          0.473            3           84          938
                                        ........  ........  Severe (heart block).....          5     ............  ...........        1,247      455,883
                                        ........  ........  Death....................          0.1   ............  ...........  ...........    5,000,000
Ephedra (undeclared)..................         1         1  Cardiovascular...........         14     ............  ...........        1,415        3,530
                                        ........  ........  CVS w/chronic............          2     ............  ...........        2,591      457,227
                                        ........  ........  Nervous system...........         14            0.47             2        1,331        1,900
                                        ........  ........  NS w/chronic.............          2     ............  ...........        2,507      455,597
                                        ........  ........  Liver impairment.........          4     ............  ...........          168        4,342
                                        ........  ........  Exfoliative dermatitis...          7     ............  ...........           84        1,206
                                        ........  ........  Other....................         54            0.29             1            0          174
                                        ........  ........  Death....................          3     ............  ...........        2,507    5,002,507
Lactose (undeclared) intolerance......         2         1  Mild.....................        100            0.48             1            0          290
Iron poisoning........................         2         1  Mild.....................        100            0.48             1           84          374
Sulfites (undeclared).................         1         1  Mild allergic reaction...        100            0.44             2            0          529
--------------------------------------------------------------------------------------------------------------------------------------------------------

    We used the transformed value of statistical life to estimate the 
value of QALD. For the most likely value of a statistical life day, we 
used $630. We derived this value from a widely-used estimate of the 
value of a statistical life: $5 million. The $5 million estimate is 
based on calculations matching labor market risks with wages for risky 
jobs. Workers in risky jobs tend to receive increased wages to 
compensate them for (usually) small increases in the probability of 
death. The implicit value of a statistical life is the increased wage 
divided by the increased probability of death. The advantage of valuing 
statistical lives with this method is that it reflects the observed 
willingness of workers, and by inference, of the whole population of 
adults, to accept small risks to their lives in a real world risk-
dollar tradeoff.
    We turn the estimated value of a statistical life into a value of a 
statistical life day by first assuming that the workers have a 
remaining life expectancy of 36 years (Ref. E18). Using a 3 percent 
social rate of time preference, the present value of 36 years is 21.83 
years. The social rate of time preference is the average long-term real 
rate of interest, with no premiums for risk and other factors that 
affect interest rates. Most analysts use the average real rate on long-
term treasury bonds (3 to 5 percent in recent years) to represent the 
social rate of time preference. The discounted expected days lost for a 
statistical death is 21.83 x 365 = 7,968. Therefore, the value of a 
statistical day is $5 million/7,968, which is approximately $630. We 
use this value to estimate the public health benefits from preventing 
illness.
    In addition to lost productivity and pain and suffering, illness 
caused by supplement contamination leads to direct medical costs. 
Direct medical costs include the cost of medicine, hospitalization, and 
visits to physicians and other professionals. We included all estimated 
medical costs, not just out-of-pocket expenses. These full medical 
costs often are missed because most medical care is covered by health 
insurance that separates the bearer of the medical cost (society) from 
the bearer of the utility losses (the ill person).
    The total costs of illnesses caused by the contamination of dietary 
supplements from poor manufacturing practices would be the costs per 
illness (classified by severity) multiplied by the number of illnesses 
(classified by severity). For chronic illnesses, the utility losses and 
medical costs stretch indefinitely into the future. We used a real 
discount rate of 7 percent to calculate the present value of chronic 
medical expenditures and utility losses. OMB suggests using a real 
discount rate of 7 percent to analyze the costs and benefits of 
regulations. This rate approximates the marginal rate of return on an 
average investment in the private sector in recent years. We used a 
different discount rate for the social rate of time preference (3 
percent) and the discount rate of future medical costs (7 percent). 
Medical costs, like all expenditures, reflect the foregone benefits 
from alternative investments. The pure social rate of time preference 
can differ from the return on private investments.
6. Benefits and Costs
    Changes in current practices by manufacturers, or consumers, or 
both, cause incremental (marginal) benefits and costs. There are 
several possible reactions manufacturers might have to the proposed 
regulatory requirements:
    ? Stop producing dietary supplements and possibly go out of 
business.
    ? Move production to a foreign country where compliance with 
these regulations is more difficult to enforce.
    ? Comply with part or all of the proposed regulation. 
Consumers will likely be confronted with higher priced dietary 
supplements but also products that are, on average, more uniform and 
higher quality. To the extent that the latter is unknown to consumers, 
they will probably reduce consumption of dietary supplements, perhaps 
in some cases substituting them with alternative products such as 
foods.
    The benefits from the proposed regulation and the regulatory 
options result from reducing contamination and adopting practices that 
will result in consistently high quality dietary supplements. Creating 
industry-wide minimum requirements for good manufacturing practices 
should reduce the occurrence of product defects, which in turn should 
reduce the number of illnesses and deaths.

[[Page 12231]]

Defective products can cause isolated cases of illnesses, but also rare 
catastrophic events such as the outbreak of eosinophilia myalgia 
syndrome (EMS) that resulted from the consumption of contaminated L-
Tryptophan. That outbreak caused 38 deaths and over 1,500 illnesses.
    The provisions that require establishments to maintain consumer 
complaint files related to manufacturing practices will generate 
additional health benefits. The use of these files by manufacturers and 
the agency will help identify dietary supplements that were 
manufactured or contaminated in ways that could cause a significant or 
unreasonable risk of illness or injury. These records may reduce the 
likelihood of catastrophic events, because a cluster of illness 
complaints could be identified, and preventive action taken before the 
number of illnesses reached catastrophic levels.
    Improved product quality will also reduce the number of products 
recalled. Certain manufacturing practices, such as more frequent 
finished product quality testing, help establishments to identify 
problems before the products are released for consumption. If defective 
products are caught before they are released, they will not be 
recalled.
    Creating minimum requirements should also generate benefits for 
consumers by reducing the variation in product quality. Creating 
verifiable minimum manufacturing requirements reduces the private 
effort necessary to distinguish products manufactured, packaged, and 
held using good practices from those using poor practices. Reducing the 
effort needed to find products with the identity, purity, strength, 
quality, and composition, among other characteristics, creates a 
potentially substantial, though implicit, benefit for consumers.
    The benefits from the proposed rule, then, are from:
    ? Reduced health costs caused by the reduced number of 
illness;
    ? Fewer product recalls, and;
    ? Greater assurance of consistent and better quality 
products.
    a. Reduced illnesses. The proposed regulation would improve the 
safety of dietary supplements, which would reduce the number of 
illnesses and the probability of deaths caused by manufacturing 
problems. The proposed rule would also improve product safety through 
the provisions requiring records and investigations of consumer 
complaints related to manufacturing practices. We assumed that the 
proposed rule would reduce both sporadic illnesses and catastrophic 
outbreaks. We estimated the reduction of sporadic or annual illnesses 
by using the agency's recall records as evidence of possible illnesses; 
class 1 and class 2 recalls of dietary supplements all involved 
adulterated products that could have caused illness if ingested. We 
estimated the reduction of illnesses from preventing catastrophic 
events by using the public health effects of the outbreak of EMS that 
resulted from consumption of contaminated L-Tryptophan.
    i. Reduced illnesses estimated from recall data. For annual 
illnesses, we used this formula for estimating the benefits from fewer 
illnesses:
Marginal health benefits = baseline (or current) number of illnesses 
caused by poor manufacturing practices x expected reduction in the 
number of illnesses brought about by the proposed regulation x health 
cost saved per prevented illness.

    We estimated the annual expected health benefits for the proposed 
rule by taking the values in table 8 of this document and weighing them 
by their incidence in the table. We computed the expected health 
benefits from preventing a single illness (of any type) associated with 
a class 1 recall as a weighted average of all potential illnesses (see 
table 8 of this document), with the potential illness divided by the 
total number of class recalls.
    The following formulas show how we calculated the average health 
benefits of preventing a single illness associated with a class 1 
recall.

$healthij = (QALD x days x $ per QALD)ij + $ 
medicalij
EBj = [Sgr]i (fij x 
$healthij)
EB [c1]
= [Sgr]j (wj x EBj)
wj = rj/([Sgr]j rj)

Where:

$healthij = health costs of severity level i of illness j;
QALD = quality adjusted life day;
$ per QALD = dollar value of a statistical day;
$ medical = direct medical costs;
Ebj = expected health benefit from preventing a single case 
of illness j;
fij = frequency of severity i of illness j ([Sgr]
fij = 1);
m = number of levels severity for illness j;
EB [c1], EB [c2]
= expected benefits from preventing an average illness 
associated with a class 1 recall or a class 2 recall;
wj = weight of illness j;
rj = number of product recalls for hazard j;
n = number of hazards or potential types of illness.

We then repeated the procedure for class 2 recalls and the associated 
illnesses in table 8 of this document. Table 9 of this document shows 
the average value of preventing a single illness associated with class 
1 and class 2 recalls.
    We estimated the annual marginal health benefits as the health 
benefits per illness for each class of recall multiplied by the 
estimated number of recalls.

Health Benefits = (EB[c1]
x estimated annual number of class 1 
illnesses prevented) + (EB[c2]
x estimated annual number of class 2 
illnesses prevented).

    To estimate the number of illnesses prevented, we started with the 
average annual number of products recalled for the decade 1990 to 
1999--six class 1 and seven class 2. As discussed above, we then 
assumed that these recalled products represented proxies for about 1 
percent of all illnesses caused by these problems leading to the 
recalls. With that assumption, we get 600 illnesses from class 1 
recalls and 700 illnesses from class 2 recalls (see table 9 of this 
document).\6\
---------------------------------------------------------------------------

    \6\ We used a probability distribution to represent the 
uncertainty associated with the number of illnesses. We modeled the 
number of illnesses prevented for each class as the average number 
of recalled products plus a negative binomial distribution 
representing unknown cases. The negative binomial distribution 
estimates the number of failures (unknown cases) that will occur 
before some number of successes (known cases) for a given 
probability of success. In the negative binomial distribution, we 
assumed that the number of recalled products were reported cases and 
that the probability of reporting equaled 1 percent (Ref. E16). The 
result is that the mean estimated number of illnesses is 100 times 
the reported number of recalls.
---------------------------------------------------------------------------

    Table 9 of this document shows the estimated value of the health 
benefits from the proposed rule using class 1 and 2 recall data.

               Table 9.--Health Benefits Using Recall Data
------------------------------------------------------------------------

------------------------------------------------------------------------
Total number of illnesses prevented, recall base.............      1,300
Total number of illnesses associated with class 1 recalls....        600
Total number of illnesses associated with class 2 recalls....        700

[[Page 12232]]

Dollar estimate of health benefit for preventing an illness      $60,000
 associated with a class 1 recall............................
Dollar estimate of health benefit for preventing an illness       $5,000
 associated with a class 2 recall............................
Dollar estimate of annual health benefits, recall base               $39
 (million)...................................................
------------------------------------------------------------------------

    ii. Health benefits from preventing a rare catastrophic event. We 
estimated the marginal health benefits from reducing the probability of 
a catastrophic event as follows:

Marginal health benefits = Change in probability of rare catastrophic 
event caused by poor manufacturing practices brought about by the 
proposed regulation x the number of illnesses caused by the rare event 
x health cost saved per illness.

    In 1989, there was a widespread outbreak of EMS resulting from 
consumption of contaminated L-Tryptophan. More than 1,500 cases (175 
acute illnesses and 1,287 chronic illnesses) and 38 deaths were 
identified in 50 states (Refs. E21 and E22). The outbreak prompted a 
recall of all dietary supplements that contained more than 100 mg per 
daily dose, which later was expanded to almost all products containing 
L-Tryptophan. We used the public health cost of this event as an 
estimate of the cost of a future rare catastrophic event associated 
with dietary supplements.
    EMS is characterized by severe myalgia and elevated eosinophils 
counts. Some of the most common symptoms are fatigue, weakness, fever, 
and arthralgia. Although a repeat of the EMS outbreak is not expected, 
it is an example of the rare, catastrophic events that should be 
prevented or mitigated by the proposed CGMP regulation. The testing 
provisions of the proposed regulation should reduce the probability 
that contaminated ingredients would be released to the public. The 
provisions for keeping complaint files and investigating complaints 
would allow more rapid identification of a major health event; the 
defective products could be identified and withdrawn well before the 
event claimed as many victims as L-Tryptophan.
    To estimate the benefits from preventing reduction in the 
probability of a rare catastrophic event occurring, we first estimated 
the period between now and the last rare catastrophic event, 1989, and 
we needed to make baseline assumptions about the likely time interval 
between events. The last catastrophic event occurred over 13 years ago, 
so we assumed that the lower bound would be 50 years. For lack of data, 
we then assumed a uniform probability distribution between these two 
bounds, which leads to a rough estimate of once in 30 years. We do not 
know how likely rare events are, nor do we actually know the likelihood 
of reducing these events by the proposed regulation. There can be no 
conclusive empirical support for the likelihood of a future event 
because the past may not predict the future in the absence of a stable 
frequency distribution that reflects a statistically significant number 
of similar events. All we know is that such an event occurred at least 
once in the recent past, and remains a possibility. We recognize that 
our lack of information about such events creates significant 
uncertainty about the social costs of these events and the health 
benefits from reducing their impact. Our estimate is meant to convey 
the potential or hypothetical enormity of such an event, not the 
certainty of such an event. We would like comments regarding our 
estimate of such an event.
    The health cost of the EMS outbreak was large because of the 
number, severity, and duration of the cases. One followup study (Ref. 
E21) found 88 percent of EMS patients were still symptomatic 21 to 64 
months after onset. The symptoms associated with EMS also frequently 
lead to activity limitations. Another study of victims (Ref. E22) found 
that 74 percent of symptomatic EMS sufferers were limited in their 
functions 12 months after the onset of illness.
    To find the health cost of the outbreak, we estimated the cost of 
the following health outcomes: Death, acute illness only, chronic 
illness with no activity limitation, chronic illness with mild activity 
limitation, chronic illness with moderate limitation, and chronic 
illness with severe limitations. To determine the cost for each of 
these health outcomes, we multiplied the lost quality-adjusted life 
days over the duration of the illness by the value of a life day. For 
medical costs, we estimated the cost of hospitalization for the EMS 
patients who required hospitalization (32 percent of all victims), by 
assuming 3 days per hospital stay. We used $1,284 as the cost per day 
of time spent in a hospital (Ref. E17). We assumed that chronic 
sufferers visited the doctor once a year at a cost of $84 per visit. We 
estimated the total cost of the event to be about $2 billion. Most of 
the cost of the outbreak comes from the deaths and severe chronic 
illnesses. Table 10 of this document shows the values used in the 
calculation. Note that the categories are not mutually exclusive. The 
average age of victims was about 50, so the value of statistical life 
was adjusted accordingly. If the event occurs about once in 30 years in 
the absence of the proposed rule, then the expected average annual cost 
would be about $66 million.

   Table 10.--Health Benefits From Preventing Rare Catastrophic Event
------------------------------------------------------------------------
                                                              Costs per
                                                   Number        case
------------------------------------------------------------------------
Hospitalization...............................          480       $3,741
Death.........................................           38    4,214,301
Acute Illness.................................          175        8,760
Chronic illness not limited...................          380    1,091,849
Mild chronic illness, limited.................          190    1,349,002
Moderate chronic illness, limited.............          307    1,601,539
Severe chronic illness, limited...............          409    1,602,844
Visits to physicians..........................        1,287        1,539
------------------------------------------------------------------------

[[Page 12233]]

    The benefits attributable to this proposed rule from preventing a 
rare catastrophic event are highly uncertain. We do not know if such an 
event would, in the absence of the proposed regulation, ever occur 
again. The EMS outbreak may have been a unique event, although the 
recent severe public health effects associated with aristolochic acid 
in Europe show that such similar events remain possible (Ref. E23). We 
also do not know that if another catastrophic event occurred, the 
health effects would be as large as for L-Tryptophan. Some of the 
smaller clusters associated with dietary supplements could represent 
small events potentially prevented by the proposed CGMP regulations 
(Ref. E15).
    We included reducing the likelihood of a catastrophic public health 
event as a benefit of the rule because the battery of checks and 
controls that would be required under the proposed regulation would 
reduce the likelihood of such an event occurring again. In particular, 
the requirement that establishments keep records of consumer complaints 
should lead to early identification and prevention of potential 
catastrophic events related to manufacturing practices.
    Our estimate of the health benefits associated with this proposal 
is based on two models that estimate future illnesses and deaths 
prevented by this proposed rule: Illnesses caused by sporadically 
adulterated products and predicted by recall data; and rare 
catastrophic outbreaks of illnesses, as predicted by one previous event 
in the United States and corroborated by one in Europe. The frequency 
and magnitude of a rare catastrophic event is largely hypothetical. In 
contrast, sporadic illnesses are small but frequent events that happen 
routinely. Small sporadic events are characterized by significant 
underreporting primarily because of the difficulty linking an illness 
with the cause of an illness. Determining the cause of an illness in 
small sporadic events is made even more difficult because only the most 
serious illnesses are likely to be reported and because of the 
difficulty of linking the cause of an illness with poor manufacturing 
practices. Catastrophes are large but infrequent events that create 
hundreds of illnesses with reporting that is close to complete because 
the public health system typically devotes considerable care in 
identifying the origin and magnitude of the problem. Adding these two 
models should not lead to double counting the health benefits. Double 
counting would most likely occur if a recalled product caused both 
sporadic illnesses and a catastrophic number of illnesses and the 
public health system accurately recorded the full number of both 
sporadic and catastrophic illnesses.
    b. Fewer products recalled. Implementation of the proposed 
regulation would reduce the number of adulterated products distributed 
to the public, which would reduce the number of products recalled. 
Manufacturing practices, such as testing of finished products and 
better recordkeeping, will increase the ability of establishments to 
identify problems before products are released for distribution. If 
adulterated products are caught before they are distributed, they will 
not be recalled.
    To estimate the direct benefits from fewer recalled adulterated 
dietary supplements, we estimated the baseline number of annual recalls 
of dietary supplements due to contamination before the proposed 
regulation. From 1990 to 1999, FDA received reports on an average of 20 
recalls per year (Ref. E12). The average figure reported here includes 
class 3 recalls. The number of units of dietary supplements for each 
recalled product varied, so we used a distribution per recalled product 
of 1,000 units to 34,000 units (Ref. E12). Product price also varied, 
with most prices falling between $5 per unit and $9 per unit; we used a 
most likely price of $7.70 per unit. We also included an adjustment for 
the goodwill lost by the establishment as a result of the recall. 
Studies of changes in market valuations of firms after recalls indicate 
that the value of lost customer goodwill, based on the decline of the 
share price of publicly traded stocks from recalls is often as large as 
the cost of the recall itself (Ref. E24). We multiplied the direct cost 
of the recall by two in order to include the lost goodwill. The result 
is an estimated savings of about $3 million per year.
    We based the estimated benefits from fewer recalled products on our 
recall data. If there were private recalls due to contaminated 
supplements that were not included in our data, the benefits from 
reduced recalls may be understated.
    c. Reduced hypothetical search costs as a measure of the benefit 
from increased assurance of quality. Consumers incur a cost if they 
purchase products but do not get the quality of product they 
anticipated. Determining the cost they incur is difficult, because we 
cannot look at the price of poor quality products and conclude that 
consumers paid too much, even when they did not get the quality they 
anticipated. We cannot disentangle the price consumers are paying, from 
the price they should be paying, because we assume consumers expect 
some unknown number of their products may not meet their expectations 
but purchase them anyway. In other words, we cannot rule out the 
possibility that the purchase price already incorporates the 
expectations of consumers that some products will be ``lemons.'' 
Because we cannot look into the minds of consumers to determine their 
expectations or their willingness to pay for these products, we can 
only estimate the benefits from more uniform quality by estimating the 
changes in behavior that would occur if consumers were aware of the 
change in quality brought about by the proposed rule. In other words, 
we assume that if the quality attributes of dietary supplements were 
observable, then consumers would spend time searching for those 
attributes, as they do for other goods. We measured this benefit as a 
reduction in the hypothetical search costs for product quality, meaning 
the identity, quality, purity, strength, and composition claimed on the 
label.
    The hypothetical measure of quality starts by assuming the 
existence of a baseline amount of search necessitated by the existence 
of poor manufacturing practices. Our hypothetical consumers must search 
for products made with good manufacturing practices, because they 
cannot take such practices for granted when purchasing dietary 
supplements. Although the search we use as a measure of the benefits 
from improved quality is hypothetical, the values we use in estimating 
our search model are based on data and inferences about real searches 
for other products.
    To get the products they want, people search across the range of 
market alternatives. Several recent articles have noted the large 
variation in product quality for different goods and services (Refs. 
E25, E26, and E27). Searching takes time and resources that could be 
used for other purposes, so a regulation that reduces search provides 
measurable benefits to consumers. To reduce the effort devoted to 
searching, consumers of dietary supplements should therefore be willing 
to pay some amount. We lack, however, a measure of what they would be 
willing to pay, partly because some consumers may not know that dietary 
supplements may contain more or less (or something not even expected) 
of what they think they are buying. Indeed, if consumers of dietary 
supplements could determine the quality of these products by merely 
examining the product or the label, the market alone would be 
sufficient to ensure that firms responded to consumer preferences for 
product quality. Consumers would search for those brands that are more

[[Page 12234]]

likely to have the desired quality, and manufacturers would most likely 
adopt sufficient quality controls to satisfy consumer preferences. The 
market response is weak now because only some consumers know that 
product quality problems exist, and even these consumers must rely on 
imperfect information. If there were uniform quality control practices 
throughout the industry that ensured against product quality defects, 
consumers would not have to search for the products that they believe 
are free from contamination or have the identity, purity, strength, 
quality, and composition they want. Consumers could more reasonably 
assume that all products are free from contamination and have the 
identity, purity, strength, quality, and composition stated on the 
label.
    We faced the problem of trying to measure what people would pay for 
more uniform products quality if they knew that manufacturing quality 
requirements did not already exist. To estimate what people would pay, 
we start with the hypothetical behavior of people aware of the lack of 
uniform product quality; we call these hypothetical people the 
``sophisticated consumers.''
    Sophisticated consumers spend time searching for signals about the 
quality of dietary supplements. The proposed CGMP regulations would 
reduce the amount of search (by some uncertain amount) carried out by 
these consumers. The benefits of the rule, however, would not be 
confined to sophisticated consumers. We also expect ``naive consumers'' 
to enjoy the benefits. Naive consumers would incur the costs of 
additional search once the correct or adverse information about quality 
is available, suffer from worry or an illness from taking poor quality 
products, or incur the cost of paying for products that do not meet 
their needs (Ref. E28). Once good practices are in place they would 
avoid these costs. Naive consumers are those who fail to search for 
quality or search little not because they do not care but because they 
do not know that quality varies as much as it does. In other words, 
they lack the information that problems exist; if they know about the 
problems, they would search or be willing to pay more to ensure that 
supplements they consume meet minimum quality standards. Although these 
naive consumers may not change their behavior in response to the 
proposed CGMP regulation, they would nonetheless enjoy the benefits. 
The naive consumers, of course, also represent real consumers of 
dietary supplements. The total benefits of the quality standards part 
of the proposed rule will be the implicit value of the gain in product 
quality enjoyed by all consumers.
    The problem is to measure that gain based on hypothetical searches. 
We needed to use data from searches in other markets, because we found 
no information on direct or indirect searching for minimum dietary 
supplement quality standards. For the sophisticated consumer, we 
assumed that the value of search time should be approximately the same 
as the willingness to pay for an attribute of the good. Sophisticated 
consumers will hypothetically search until the expected benefit of 
continued searching is less than the expected cost of continued 
searching. The total cost of search time will, on average, be no more 
than the expected cost of the additional quality desired. Search time 
includes the time spent: Reading product labels and other literature 
about the product, comparing one product with other products, examining 
the product itself (sometimes carefully), thinking about the product, 
and second guessing final decisions. It might also include the time 
actually shopping for the product: Finding the locations where the 
product is sold, driving there and back, waiting in checkout lines, and 
walking up and down the aisles.
    We used information on shopping times for a range of products to 
derive an estimate for the hypothetical search time for dietary 
supplements. We assumed that some fraction of shopping time is pure 
search time, although we also recognize that search time includes more 
than the search for product quality. Some search time, for example, is 
for price, efficacy, and other attributes. The reduction in search time 
for the sophisticated consumer would therefore be at most a fraction of 
total search time for dietary supplements. The measure of time saved 
then is:

Reduced search time due to CGMP regulation = shopping time x fraction 
of shopping time spent searching x fraction of search time associated 
with searches for quality x fraction of search time associated with 
searches for quality that would be eliminated if CGMP rule guaranteed 
minimum quality.

    We took the estimated reduction in hypothetical search time for the 
sophisticated consumer and applied it to all consumers to get an 
estimate of the implicit benefits of establishing minimum quality 
standards. This estimated saving in hypothetical search time is not a 
forecast of reduced shopping time; it is a proxy measure of the benefit 
from reduced variance and improved mean product quality. We anticipate 
little or no change in aggregate shopping time for dietary supplements.
    We converted the time measure into a monetary measure by 
multiplying the time reduction for sophisticated consumers by the 
average wage rate. The benefits measure reduced search time associated 
with improved quality assurance:

Quality assurance benefits = reduction in search time (in hours per 
year) per sophisticated consumer x average wage rate per hour x total 
number of consumers.

The shopping time model is an indirect approach to measuring benefits 
in a market with asymmetric information; it is not a prediction about 
how shopping behavior will change in that market. Indeed, we believe 
that most of the beneficiaries of this part of the rule will never 
recognize that they are beneficiaries.
    Standardization imposes minimum requirements on manufacturing, 
which in turn should reduce the variance of product quality. The 
reduction in product quality variation should reduce the amount of 
information sophisticated consumers need to acquire before purchasing 
dietary supplements (Ref. E29). People need not rely as much on such 
indicators as brand names, price, place of purchase, articles in 
consumer magazines, or advertising to determine the likelihood that 
dietary supplements meet minimum quality standards.
    Although no studies deal with dietary supplements directly, the 
literature on consumer search for other commodities provides insights 
that increase our understanding of the search costs for supplements 
(Refs. E30 and E31). Duncan and Olshavsky (Ref. E32) surveyed buyers of 
television sets and found that 88 percent of respondents performed some 
type of search activity before purchase. In a study (Ref. E33) of 
consumer search for microwave ovens, the average buyer of a new 
microwave oven was willing to search for four alternative products. 
Search for groceries has been characterized as a two-stage process 
(Ref. E34). First, people engage in prestore activities, such as 
reading advertisements, writing shopping lists, clipping coupons, and 
comparing stores. Second, people engage in search activities at the 
store, including price and product comparison and search for items with 
coupons. Most people devote time to search activities for all but the 
most routine purchases.
    To estimate the reduction in hypothetical search costs from the 
proposed rule, we started with estimates

[[Page 12235]]

of the time consumers spend in search for groceries and other household 
purchases (including durable goods). We assumed that the search time 
for these products was related to shopping time. Because search costs 
include the costs of evaluating magazine articles or brochures, the 
costs of obtaining a friend's advice, and the costs of instore product 
comparisons, our estimates will not correspond precisely to the actual 
costs of search for these products (Ref. E35). We believe, however, 
that the measure will be a reasonable approximation. Although search 
time often takes place outside of measured shopping time, measuring 
search time as some proportion of total shopping time should generate a 
plausible if not a precise estimate.
    We generated three models of search time for dietary supplements, 
based on three separate studies of shopping time:
    ? Drug Store.
    ? Use of Time.
    ? Grocery Store.
    We used three models based on different assumptions because using a 
range of studies reduced the likelihood of systematic bias in our 
analysis.
    The drug store model. The drug store study recorded the amount of 
time people spent looking at an item on the shelf before making a 
purchase (Ref. E36). Customers, on average, spent 3.75 minutes studying 
a product before purchasing it. Although there are quality standards in 
place for over-the-counter drugs and not for dietary supplements, we 
assumed that this represented a measure of the amount of time the 
sophisticated consumer might spend searching for a product with the 
desired quality.
    The use of time model. The Americans' Use of Time Project (Ref. 
E37) used time diaries to study how adults spent all of their time. The 
study collected data from over 3,500 adults on use of time. Data from 
these time diaries reveal that adult Americans spent about 364 minutes 
per week shopping for personal consumption items, such as groceries and 
other household products.
    The grocery store model. In the grocery store study, hidden 
observers tracked and recorded shopping time in the store (Ref. E38). 
The study found that people on average spent about 21 minutes shopping 
in the grocery store. By combining estimated time per trip with the 
Food Marketing Institute's (Ref. E10) finding that consumers average 
about 2.2 grocery shopping trips per week, we generated an estimate of 
search time for all grocery store purchases of 46.2 (= 2.2 x 21) 
minutes per week.
    For each of the models, we needed to make assumptions to convert 
shopping time for other commodities into search time for dietary 
supplements. Table 11 of this document shows the assumptions and 
information used in each model.

 Table 11.--Three Models of Search Time: Assumptions Used in Simulations
------------------------------------------------------------------------
                                    Value or
          Variable                distribution        Source and notes
------------------------------------------------------------------------
                            Drug Store Model
------------------------------------------------------------------------
Search time in minutes per    3.75................  Ref. E30.
 item.
Number of products per        6.57................  Ref. E4.
 person per year.
Average wage rate...........  $15.65 per hour, or   Ref. E42.
                               $0.26 per minute.
Population..................  273 million.........  Ref. E19.
Fraction of search time       0.2 (based on         Based on number of
 devoted to searching for      uniform               attributes
 quality.                      distribution, 0.1     consumers search
                               to 0.3).              for.
-----------------------------
                            Use of Time Model
------------------------------------------------------------------------
Weekly shopping time for all  346.................  Ref. E37.
 items in minutes.
Fraction percent of budget    $15.5 billion/$6,250  Ref. E4 and E19.
 spent on supplements.         billion.
Average wage rate...........  $15.65 per hour, or   Ref. E42.
                               $0.26 per minute.
Adult population............  205 million.........  Ref. E19.
Ratio of search time to       0.7 (based on         Based on
 shopping time.                uniform               descriptions of
                               distribution, 0.4     shopper behavior.
                               to 1.0).
Fraction of search time       0.2 (based on         Based on number of
 devoted to searching for      uniform               attributes
 quality.                      distribution 0.1 to   consumers search
                               3.0).                 for.
Potential reduction in        33% most likely       Based on likelihood
 search time attributable to   (could be between     of problem and
 CGMP regulations.             15 and 50%).          likelihood that
                                                     search will decline
                                                     proportionally, and
                                                     the expert opinion
                                                     of pharmacists.
-----------------------------
                           Grocery Store Model
------------------------------------------------------------------------
Weekly shopping time for      46.2................  Ref. E38.
 groceries in minutes.
Ratio of supplement           $15.5 billion/$710    Ref. E38.
 expenditures to grocery       billion.
 expenditures.
Average wage rate...........  $15.65 per hour, or   Refs. E4 and E19.
                               $0.26 per minute.
Adult population............  205 million.........  Ref. E19.
Ratio of search time to       0.7 (based on         Based on
 shopping time.                uniform               descriptions of
                               distribution, 0.4     shopper behavior.
                               to 1.0).
Fraction of search time       0.2 (based on         Based on the number
 devoted to searching for      uniform               of attributes that
 quality.                      distribution, 0.1     consumers search
                               to 0.3).              for.
Potential reduction in        33% most likely       Based on likelihood
 search time attributable to   (could be between     of problem, the
 CGMP regulations.             1% and 50%).          likelihood that
                                                     search will decline
                                                     proportionally, and
                                                     the expert opinion
                                                     of pharmacists.
------------------------------------------------------------------------

    The drug store data generated a direct estimate of search time. In 
the drug store model we assumed that the time spent standing in front 
of the drug product could be used to estimate the time searching for 
dietary supplements. We then used data on the number of products 
purchased per person and the total U.S. population to generate an 
estimate of annual search time for dietary supplements.

[[Page 12236]]

    To estimate the time spent searching for supplements from the use-
of-time study, we assumed that the share of all shopping time devoted 
to supplements would be proportional to the share of a consumer's 
budget spent on supplements. We recognize that it could well be higher 
if supplements require more search than the average commodity. 
According to an industry source and FDA projections, consumers spent 
about $15.5 billion on dietary supplements in 1999 (see table 5 of this 
document). Consumers spent about $6,250 billion on all personal 
consumption in 1999, which means that dietary supplements accounted for 
about 0.24 percent of those expenditures. Personal consumption 
expenditures included in this estimate are food, alcoholic beverages, 
housekeeping supplies (such as laundry and postage), household 
furnishings and equipment (such as furniture and appliances), apparel 
(includes footwear), personal care products and services, reading 
materials, tobacco products, and smoking supplies. Annual shopping time 
per person for dietary supplements would therefore be about 44.6 
minutes per year (= ($15.5 billion/$6,250 billion) x 346 minutes per 
week x 52 weeks). We converted shopping time to search time by assuming 
that search time equaled 40 to 100 percent of shopping time. Total 
search time equaled search time per adult multiplied by 205 million 
adults. We assumed that all adults would perform search, although we 
recognize that not all adults consume dietary supplements and not all 
search is conducted by adults. Children might search for these products 
also. The opportunity cost for children, as measured by their wage rate 
is much less than for adults, so we assumed their search time could be 
ignored. We used the total adult population rather than just the adult 
consumers of dietary supplements, because the shopping time studies are 
for all adults.
    We estimated search time in the grocery store model with 
assumptions similar to those in the use-of-time model. We assumed that 
the ratio of search time for supplements to search time for groceries 
would equal the ratio of expenditures on supplements to expenditures on 
groceries. Estimates from the 1998 Consumer Expenditure Survey (Ref. 
E39) (adjusted for changes in prices between 1998 and 1999) reveal that 
consumers spent approximately $710 billion on grocery store purchases 
in 1999. Grocery store purchases included food, alcoholic beverages, 
housekeeping supplies, personal care products, tobacco products, and 
smoking supplies. Annual shopping time per person for dietary 
supplements would therefore be about 52.5 minutes per year (= ($15.5 
billion/$710 billion) x 46.2 minutes per week x 52 weeks). We again 
converted shopping time to search time by assuming that search time 
equaled 40 to 100 percent of shopping time. Like the estimate from the 
use of time model, this value was then multiplied by 205 million 
adults.
    We used these three models based on different assumptions because 
we wanted to explore a range of studies to avoid systematic bias in our 
analysis. We recognize that the three estimated annual search times for 
dietary supplements do not represent the search for quality alone. 
Consumers search for a variety of features; only part of every search 
will be devoted to quality. We assumed that 10 to 30 percent of pure 
search time involves quality searches. Estimating the impact of CGMP 
regulations on consumers' search time is difficult, since no previous 
studies have analyzed the changes in search time following the adoption 
of CGMP regulations or from increases in product quality 
standardization. However, a consistent finding from the literature is 
that search time should decline following a decrease in the variation 
in product quality (Refs. E35 and E40). In the absence of previous 
empirical studies, we assumed that the proposed rule would reduce the 
hypothetical search time for quality ``the search time of sophisticated 
consumers'' by 1 to 50 percent, with 33 percent the most likely value. 
A survey of pharmacists reported their belief that 30 percent of their 
customers place manufacturing quality as a top priority in selecting 
one herbal over another (Ref. E41). We also used evidence from product 
tests that indicated that up to 33 percent of products were missing key 
ingredients or contained unwanted ingredients (Refs. E25, E26, and 
E27). If the proposed rule guarantees that products will contain what 
the label claims, then perhaps search time for quality will decline by 
that percentage.
    To estimate the value of the possible reduction in searching for 
quality, we multiplied our estimated time saving by the average wage 
rate, which is an estimate of the value of time. The average hourly 
wage rate for U.S. workers was $15.65.\7\ We ran computer simulations 
of all three models. The results for the three models are shown in 
table 11 of this document.
---------------------------------------------------------------------------

    \7\ Personnel Employment, Hours, and Earnings. Series ID: 
EES00510006 Seasonally Adjusted, Industry: Goods-producing Data 
Type: Average hourly earnings of production workers, Employment Cost 
Index, Bureau of Labor Statistics.
---------------------------------------------------------------------------

    d. Other benefits. The proposed regulation could also reduce the 
total time and effort that all covered establishments expend to monitor 
ingredient suppliers and holders of their products. Because all 
ingredients and holders would be subject to the same uniform minimum 
requirements, variation in their practices would decline, so firm 
monitoring of upstream and downstream vendors could decline.
    The provision that requires establishments to maintain complaints 
files would allow a manufacturer to more readily be able to identify a 
product that causes a significant or unreasonable risk of illness or 
injury. The manufacturer can then take necessary steps to prevent any 
additional adverse health impact. We have attempted to quantify this 
benefit for preventing catastrophic events, but not for reducing 
smaller risks. FDA adverse event reports, however, imply that many such 
small events occur, and the proposed rule could prevent some of them 
(Ref. E15).
    In addition, if the same adverse events show up in complaints 
received by different firms selling products with the same or similar 
manufacturing problems, no one firm selling such products may recognize 
the need to investigate the complaints especially if the risk is 
relatively low. Because we would have access to complaint files, our 
review would be more likely than any individual firm's review to 
identify the need to investigate the complaint because of a reasonable 
possibility of a relationship between the manufacturing process of a 
dietary supplement and the adverse event.
    e. Total measured benefits. The total measured benefits from the 
proposed rule are the sum of the value of health benefits, the value of 
the reduced number of product recalls, and the reduction in 
hypothetical search costs. Table 13 of this document shows the total 
benefits.

        Table 12.--Three Models to Estimated Search Cost Savings
------------------------------------------------------------------------
                                                           Cost savings
                     Baseline model                       (in millions)
------------------------------------------------------------------------
Drug store model.......................................             $108
Use of time model......................................              101
Grocery store model....................................              119
Average of three baseline models.......................              109
------------------------------------------------------------------------

[[Page 12237]]

                  Table 13.--Summary of Annual Benefits
------------------------------------------------------------------------
                                                            Mean  (in
                        Benefits                            millions)
------------------------------------------------------------------------
Fewer illnesses (from table 8).........................              $39
Fewer illnesses (from table 10)........................               66
Fewer product recalls (from table 9)...................                3
Reduced consumer search (from table 12)................              109
                                                        ----------------
    Total benefits.....................................              218
------------------------------------------------------------------------

7. Costs
    The same changes in practices that produce benefits also have 
costs, the opportunity costs of not doing what consumers and 
manufacturers are now doing. The proposed regulation would require 
dietary supplement establishments to adopt some new practices in order 
to manufacture, package, and hold their products. The costs incurred 
for those who choose to comply will be for personnel, grounds and 
physical plant, equipment and instrumentation controls, quality control 
and laboratory operations, production and process controls, handling 
consumer complaints, and holding. In some cases, establishments would 
need to make capital improvements to the physical plant, add or replace 
equipment or controls, perform additional maintenance, keep records, 
carry out tests, or execute a variety of additional tasks that they may 
not have previously performed. We estimated the additional costs of 
production associated with the proposed rule and the leading regulatory 
options, using the survey (Ref. E2) to estimate baseline manufacturing 
practices.
    a. Description of the costs. To estimate costs for the dietary 
supplement industry, we initially divided the industry into four 
product categories and three size categories. Because the survey showed 
that there were only a few establishments in some categories, we 
consolidated the size and product into three size categories. The size 
categories were:
    ? Very small (fewer than 20 employees).
    ? Small (20 to 499 employees).
    ? Large (500 or more).

    Although this consolidation glosses over the important differences 
across products, the purpose is to estimate the broad average costs of 
the rule.
    For each category, we constructed a cost model that included every 
provision of the CGMP regulations that the proposed rule requires or 
recommends. We then attached a cost to each provision that had an 
activity associated with it. Most provisions did not have costs 
attached to them, mainly because they were either descriptive or the 
costs were included elsewhere. For the rule as a whole, we estimated 
the marginal, or additional costs for over 70 provisions of the 
proposed rule.
    We expressed the cost as cost per unit, with the unit being either 
the establishment, the number of employees, or the annual number of 
batches produced. The costs of this proposed rule included the 
following general activities: Sanitation, production and process 
controls, holding and distributing, and consumer complaints.
    b. Costs of general activities. i. Sanitation. Sanitation includes 
both one-time capital improvements and ongoing efforts. Some provisions 
of the proposed regulation may require establishments to perform one-
time capital improvements to their physical plant facilities.
    The proposed regulation would also require, if not already in 
place, physical plant owners to install new or additional plumbing 
systems to carry additional water or sewage, additional toilet or hand 
washing facilities, additional facilities for trash disposal, or new 
signs to instruct employees. The proposed regulations might also 
require establishments to add space in order to keep equipment and 
materials farther apart, which will help to prevent contamination or 
mixups. Other possible capital expenditures (among many other possible 
requirements) include:
    ? Replacing floors, walls, or ceilings with smooth, hard 
surfaces;
    ? Changing fixtures, ducts, or pipes that might be a source 
of contamination by dripping or condensation;
    ? Adopting ventilation control systems including filters, 
fans, or other air-blowing equipment to prevent odors or vapors;
    ? Additional lighting to ensure that equipment, contact 
surfaces, or other areas where supplements are examined, processed, or 
held can be adequately seen.
    Sanitation also requires that equipment utensils must be of 
suitable design, construction, and workmanship to enable them to be 
adequately cleaned and maintained. To meet this requirement, some 
establishments may need to provide additional maintenance or additional 
cleaning and sanitation for their equipment and utensils. Also, 
freezers and cold storage compartments used to slow or arrest the 
growth of microorganisms must be fitted with thermometers to accurately 
show the temperature within the compartments. Instruments and devices 
used in manufacturing must be accurate, adequately maintained, and 
adequate in number. To meet this requirement establishments might have 
to purchase new equipment, replace old equipment, or provide additional 
maintenance to existing equipment.
    ii. Production and process controls. Production and process 
controls are the main preventive mechanism to ensure the identity, 
purity, quality, strength, and composition in the proposed rule. 
Establishments must implement a system of production and process 
controls that covers all stages of processing, from the receipt and 
acceptance of components, dietary ingredients, dietary supplements, 
packaging, and labels through the release for distribution and holding 
of the dietary ingredients and dietary supplements. Establishments must 
identify points, steps, or stages in the manufacturing process where 
control is necessary to prevent adulteration. Establishments must also 
establish specifications for the identity, quality, purity, strength, 
and composition of components, dietary ingredients, or dietary 
supplements. Establishments must monitor the points, steps, or stages 
in the batch production, as specified in the master manufacturing 
record, where control is necessary to prevent adulteration. 
Establishments must establish specifications for packaging to ensure 
that containers or closures that come into contact with dietary 
ingredients or dietary supplements are not reactive or absorptive and 
are composed of substances that are safe for use in or on food.
    Establishments that have not already done so must establish a 
quality control unit with one or more individuals that have with the 
authority and responsibility to review the results of monitoring, make 
decisions on the disposition of materials, and identify whether actions 
taken to correct any deviations are appropriate. The quality control 
operation must ensure that components, dietary ingredients, and dietary 
supplements conform to specifications.
    iii. Holding and distributing. Establishments must hold and 
distribute dietary ingredients and dietary supplements under 
appropriate conditions of temperature, humidity, and light so that the 
identity, quality, purity, strength, and composition of the dietary 
ingredients and dietary supplements are not affected. Establishments 
must also identify and

[[Page 12238]]

hold components, in-process materials, and dietary supplements under 
conditions that will protect them against mixups and physical, 
chemical, and microbial contamination. Packaging materials must also be 
protected against deterioration. Establishments that do not now perform 
these requirements and the other provisions associated with holding 
will incur a compliance cost.
    iv. Consumer complaints. The quality control unit must review all 
consumer complaints involving the failure of a dietary supplement to 
meet any of its specifications, or the failure to meet any other 
requirements under proposed part 111, including those specifications 
and other requirements that, if not met, may result in possible illness 
or injury. In addition, the quality control unit must investigate such 
a consumer complaint where there is a reasonable possibility of a 
relationship between the consumption of a dietary supplement and an 
adverse event. The complaint and report of the investigation results 
should be reported to FDA when there is a possibility of a serious 
adverse event.
    c. Major costs by type of activity. Within these four categories 
(sanitation, production and process controls, holding and distributing, 
consumer complaints), the major costs of the proposed rule are 
recordkeeping (except for sanitation), capital costs for physical plant 
and equipment, finished product quality testing (part of production and 
process controls only), labor costs for certain required tasks, and 
some other costs that were not easily classified.
    i. Recordkeeping. We used a study of a medical device CGMP 
regulation to estimate the costs of recordkeeping (Ref. E44). We 
request comments on the applicability of a study of the medical device 
CGMP's to dietary supplements.
    The compliance cost of recordkeeping is the sum of both the initial 
design and printing of the recordkeeping documents and the recurring 
costs of maintaining the records. The cost of training personnel to use 
mandatory records is a recurring cost that depends on how frequently 
records are modified, the frequency of personnel turnover, and how 
complicated the tasks are that are being recorded. The recurring costs 
are measured by the workers' wage rate, which we assumed is $15.65 per 
hour based on the average manufacturing wage, multiplied by the 
expected labor hours necessary to perform a written or electronic 
record and the time necessary for management to review the records to 
see the actions are documented accurately. For electronic records, the 
recurring time is the time necessary to ensure that the equipment is 
serviced and maintained properly.
    ii. Capital costs for physical plant and equipment. We estimated 
capital costs for physical plant redesign at $50 per square foot (Ref. 
E45). For establishments with inadequate facilities, we assumed that 
between 0 and 20 percent of the physical plant would have to be 
renovated, with 10 percent the most likely. For equipment costs, we 
assumed that very small establishments would on average spend 0 to 
$1,000, with $100 the most likely amount. Small establishments would 
bear costs 3 times that of very small establishments, which is the 
ratio of the size of the physical plants of small establishments to the 
size of the physical plants of very small establishments. We assumed 
that large establishments would bear (if necessary) costs 20 times that 
of very small establishments, which is the ratio of the size of the 
physical plants of large establishments to the size of the physical 
plants of very small establishments. In other words, we assumed capital 
costs for physical plant and equipment would be proportional to 
facility size, as measured in square feet.
    iii. Testing. Establishments that do not already conduct the 
required product quality tests of each batch of dietary ingredients or 
dietary supplement produced would incur the cost for those tests. Under 
the option for more restrictive CGMP rules, each lot of components 
would also be tested. The costs per establishment depend on both the 
number of tests and the costs per test. We did not estimate the cost of 
developing new, validated tests methods because we lacked information 
about the costs for this requirement and the number of such tests that 
need to be developed. We ask for comments on the costs to develop 
tests, for the number of tests and the costs for performing each test 
to comply with this requirement.
    ? Number of tests: Model. To estimate the costs of testing, 
we first estimated the number and costs of individual tests, without 
adjusting for the amount of testing already being done. In this section 
we show how we estimated the likely number of required tests, 
unadjusted for current voluntary testing. For a representative 
manufacturer, the annual number of tests would be the number of new 
tests per batch multiplied by the number of batches produced in a year.
    The proposed rule requires only tests for identity, purity, 
quality, strength, and composition of the final product. The option for 
stricter CGMP regulations would also require tests of components. 
Estimating the number of component tests per batch is complicated, 
because component tests are made on the shipment lots, rather than on 
the parts of the lots that actually go into the final product. For 
example, if a lot of some ingredient is used in 6 batches of final 
products, it would probably be tested only once.
    The establishment itself may test the shipment lots, and during 
inprocess stages for identity, purity, quality, strength, and 
composition, unless final product testing is done.
    The number of component tests per batch of final product would 
equal the number of tests per component, multiplied by the number of 
components per batch, divided by the batches per shipment lot (to 
account for the production of multiple batches of dietary supplements 
from single lots of components).
    The option for stricter CGMP regulations options would also require 
some inprocess tests upon receipt. The number of inprocess tests per 
batch is the same as the number of potential inprocess product defects. 
The estimated number of inprocess tests counts only tests for defects 
that can occur during production, not tests for the defects of dietary 
ingredients and components supplied to the producer.
    We used the following formulas to estimate the number of tests:
    [GRAPHIC][TIFF OMITTED]TP13MR03.000
    
Quality tests per batch of final product = max [m x (1/z), 1]

Where:

Ij = jth listed ingredient;
m = number of ingredients per batch;
Rj = required tests for ingredient j;

[[Page 12239]]

Uk = kth unlisted component (an inactive substance);
n = number of unlisted components per batch;
Rk = required tests for unlisted component k;
S = number of shipments (or lots) of ingredients and unlisted 
components;
B = number of batches produced;
 H1 = 1th inprocess potential defects;
R1 = required inprocess tests per batch for potential defect 
H1;
o = number of potential inprocess defects per batch;
z = number of ingredients identified per quality test.

    ? Number of tests: Evidence and distributions. The quantity 
and quality of evidence on the variables used to estimate the number of 
required tests varies greatly. In this section, we explain the evidence 
and assumptions we used to construct the formulas for the number of 
tests.
    ? Number of ingredients. We based our measure of the number 
of dietary ingredients per product on a sample of almost 3,000 dietary 
supplement labels (Ref. E46). Although some dietary ingredients may be 
missing from the labels and some listed dietary ingredients may be 
missing from the products, the ingredient list represents the best 
evidence we are likely to have on what dietary ingredients are used in 
dietary supplements.
    ? Number of ingredients per batch. According to the sample of 
listed ingredients (Ref. E46). Vitamin and mineral products contain 
about 13 listed ingredients. Other dietary supplements, mainly herbals, 
contain about four.
    ? Number of tests per ingredient lot. The option for more 
restrictive CGMP regulations would require that virtually all dietary 
ingredients be tested for identity and defects at some stage between 
harvesting the raw product and the beginning of the production of the 
final product. We assumed one identity test per ingredient lot. The 
number of tests for defects depends on the number of possible defects, 
which can include: Filth; Microbial pathogens; Chemical hazards, 
including pesticides; Insects; Physical hazards, such as metals; 
Natural toxins, such as aflatoxin; and Inadequate purity, quality, 
strength, or composition.
    The number of potential defects is potentially unlimited. As a 
practical maximum, however, few products would have more than five 
potential defects. In the calculation of ingredient testing costs (part 
of the option for more restrictive CGMP regulations), we assumed that 
the average number of tests per listed dietary ingredient would be 
between one and six: One identity test for identity, purity, strength, 
quality, and composition and zero to five tests for defects.
    ? Number of unlisted components. Dietary supplements are 
manufactured using solvents, binders, and lubricants that may not show 
up in the final product. An industry source (Ref. E47) says that four 
to six unlisted components are typical per product, although fewer are 
certainly possible. The minimum number is zero. We assumed that the 
number of unlisted components would be zero to six, with four the most 
likely.
    ? Number of tests per unlisted components. The unlisted 
components tend to be manufactured products, such as solvents. 
Therefore, one identity test would likely be sufficient.
    ? Number of shipments (or lots) of ingredients and unlisted 
components. We have no direct evidence on the number of shipment lots 
of dietary ingredients and components. We also have no evidence on the 
number of shipments per lot or on the number of shipments per batch. 
The increasing use of just-in-time inventory practices indicates that 
one shipment lot of components per batch may be the rule for some 
products and some producers. It is costly and difficult to store 
ingredients for an extended time, so establishments tend to buy more 
and smaller lots of components rather than a few large lots and storing 
them in bulk over an extended period (Ref. E48). Crude botanical and 
other ingredients are inherently unstable and may lose their quality in 
even a short time unless costly temperature, humidity, and light 
controls are in place (Ref. E49). We also know, however, that some 
dietary ingredient suppliers produce large amounts and then ship out 
smaller packages. For dietary supplements produced using part of a 
large production run of a dietary ingredient, the number of batches per 
lot could be large. Also, some producers buy a single shipment lot of a 
raw material and use it in many batches. We assume that as many as 12 
batches per shipment lot of dietary ingredient is a plausible maximum. 
In the cost calculation, we assumed that 1 was minimum and 12 the 
maximum number of batches produced per lot, with 6.5 the average.
    ? Number of batches produced. We have survey results (Ref. 
E2) on the number of batches produced per establishment. According to 
the survey, very small establishments produce an average of 223 batches 
per year, small establishments produce an average of 554 batches per 
year, and large establishments produce an average of 309 batches per 
year.
    ? Inprocess potential defects. Inprocess defects involve many 
of the same potential defects that can occur in components. The more 
restrictive CGMP option requires inprocess tests at all points where 
contamination or other defects can occur. Filth, chemicals, microbial 
pathogens, physical objects, and insects can be introduced into the 
product during manufacturing. In addition, purity, quality, strength, 
and composition can be compromised.
    ? Number of potential inprocess defects. Some processes may 
have no control points, steps, or stages that involve the potential for 
defects. If certain manufacturing processes in the production of a 
dietary supplement can be carried out without being subject to 
potential defects, no inprocess tests would be required for those 
processes. We therefore assumed that zero inprocess tests would be the 
lower bound requirement. For the upper bound, we assumed that no 
products would have more than five potential control points or steps 
that could lead to defects. We believe that most production processes 
will have fewer than 5 control points, so we assumed an average of 2.5 
control points requiring inprocess tests for defects.
    ? Number of required inprocess tests per control point. We 
assumed one test per defect per control point.
    ? Number of ingredients identified per quality test. We had 
no direct evidence on the number of identity tests per final dietary 
supplement. For the maximum, we assumed that the number of tests would 
equal the number of ingredients. The number of ingredients identified 
per test varies from less than one to a very large number. We assumed 
that for vitamins and minerals, the minimum number of identity tests 
would be one and the maximum would be 30, with 2 the most likely. 
Botanical and herbals are less easily characterized than vitamins; so 
identifying large numbers of ingredients with a single test would be 
highly unlikely. We assumed that one to two ingredients would be 
identified per test for herbal products.
    ? Number of final product tests per batch. We had no direct 
evidence on the number of quality tests per final dietary supplement. 
After adjusting for the possibility of multiple results from a single 
test, multiple ingredients in single products, and the differing number 
of ingredients in herbal and vitamin products, we estimated that the 
proposed rule would require about three tests for identity, purity, 
quality, strength, and composition for each batch of final product. 
These are the

[[Page 12240]]

only required tests in the proposed rule, but establishments may choose 
to perform inprocess tests and tests on ingredients in order to prevent 
waiting until final product testing to discover defects.
    iv. Costs per test. We estimated the costs per test partly with 
published prices of independent laboratories as posted on the Internet 
(Refs. E50 and E51), and partly from our conversations with FDA and 
industry experts on testing. We found that testing costs vary according 
to frequency and complexity. The more frequently technicians perform 
tests, the lower are the costs per test. Many tests require 
sophisticated equipment, such as gas chromatography, high pressure 
liquid chromatography, distillation, extraction, various 
spectrophotometers, and other types of equipment. Using sophisticated 
equipment requires trained personnel. Even simple physical or 
organoleptic testing requires training or experienced personnel. The 
type of ingredient, compound, or product can also affect the cost 
because some are easily identified using routine or single step 
techniques and others require multiple steps or complex techniques, 
especially if there are similar products that can be mistaken for the 
products being identified. The type of defect tested for affects the 
cost; some defects can be found visually if they are found on the 
surface, but others are latent. Some tests require multiple samples or 
multiple steps. In addition, tests require the taking and of samples, 
whose cost can vary.
    We assumed that $20 per test represented a plausible lower bound. 
This cost represents the full cost of carrying out a test, including 
collecting and storing the sample, the time for training the personnel 
who carry out the test, and any associated records. Although some 
Internet testing prices for tests were as high as $300, we assumed that 
with frequent testing $150 would be a more plausible upper bound 
average cost. The majority of listed prices fell into the $20 to $80 
range, so we selected $50 (the midpoint) as most likely. The average 
cost per test was about $60.\8\
---------------------------------------------------------------------------

    \8\ The average cost is higher than the most likely cost because 
we modeled costs with a Beta-Pert distribution that was skewed 
rightward (toward higher costs). The Beta distribution is part of 
the Bernoulli family of distributions and is closely related to the 
Binomial. The Binomial gives the distribution of the number of 
successes (s) in n trials if the probability of the success in each 
trial is p. The Beta shows the distribution of the value of p when s 
successes occur in n trials. The Beta-Pert distribution is a Beta 
distribution that has been rescaled to run between values other than 
0 and 1. The Beta-Pert uses a minimum, maximum, and most likely 
value to generate a distribution running from the minimum to the 
maximum, with a mean equal to (minimum + (4 x most likely) + 
maximum)/6. We used the Beta-Pert distribution because we did not 
have a representative sample to derive the distribution, but we did 
have enough information to identify a plausible maximum, minimum, 
and most likely value. The use of the Beta-Pert, then, indicates 
that we do not know the shape of the probability distribution of 
possible testing costs, but we do have limited data.
---------------------------------------------------------------------------

    Changing our assumption about the midpoint of testing costs would 
change our estimate of the cost of the rule. If the cost of testing 
each batch is actually significantly higher, then the impact to those 
firms that incur the cost and to society will have been understated.
    v. The number and cost of tests: summary. We estimated the number 
of tests required of the representative manufacturer as a weighted 
average of the number of tests required for vitamins and minerals and 
the number of tests required for all other supplements (which were 
mainly herbal products). We used survey responses to a question about 
the establishment's primary line of business for the weights used to 
compute the average number of tests. We dealt with multiple responses 
by treating all nonvitamin and nonmineral responses as other dietary 
supplements. The following weights, as shown below, differed by size of 
manufacturer:
    ? 24 percent of very small manufacturers produce vitamins and 
minerals; 76 percent produce other dietary supplements.
    ? 42 percent of small manufacturers produce vitamins and 
minerals; 58 percent produce other dietary supplements.
    ? 69 percent of large manufacturers produce vitamins and 
minerals; 31 percent produce other dietary supplements.
    The annual cost of testing differed by the size of the firm, 
because the average number of batches produced differed. For the option 
calling for more strict regulation, the total costs of testing would be 
much higher than in the proposed rule. The unadjusted total cost of 
testing under the more restrictive CGMP option would be:

$148,000 for very small establishments;
$415,000 for small establishments;
$263,000 for large establishments.

    We estimate that the adjusted total cost for testing for the 
proposed regulation will be:

$11,230 for very small establishments;
$19,907 for small establishments;
$7,626 for large establishments.

    We found some corroboration for these estimates in a comment on the 
Advance Notice of Proposed Rulemaking entitled ``Current Good 
Manufacturing Practice in Manufacturing, Packing, or Holding Dietary 
Supplements'' published in the Federal Register of February 6, 1997 (62 
FR 5699 to 5709). According to the comment, the cost of testing 
components and final products inhouse would be at least $650 per batch 
plus microbiological tests. Testing costs could be more if 
establishments sent samples to independent laboratories for testing or 
if they conducted extensive identity tests of herbal and botanical 
products. If we apply the $650 to the annual number of batches per 
establishment, the comment implies that very small establishments would 
perform $145,000 (223 x $650) worth of tests, small establishments 
would perform $360,000 (554 x $650) worth of tests, and large 
establishments would perform $200,000 (309 x $650) worth of tests. 
These estimates are reasonably close to our simulation estimate.
    The unadjusted testing costs represent the total requirements and 
recommendations, not the additional costs that would be incurred in 
response to the proposed rule. Tests on incoming components and 
inprocess tests would not be required by the proposed rule. Most 
establishments already conduct some tests, or send samples out for 
testing. We, therefore, adjusted the estimated testing costs of the 
proposed rule to include only required tests and to account for the 
testing costs currently borne voluntarily by manufacturers. The survey 
results showed how many respondents were conducting various types of 
tests.

           Table 14.--Values Used in Testing Cost Calculations
------------------------------------------------------------------------
                                    Value or
            Name                distribution used          Source
------------------------------------------------------------------------
Number of dietary             Vitamins and          Sample from 3,000
 ingredients per product       minerals--13; All     dietary supplement
 batch.                        other categories--4.  labels (Ref. E46).

[[Page 12241]]

Number of identify tests per  1 Identity test per   Assumption based on
 ingredient lot.               ingredient lot.       discussions with
                                                     industry--FDA
                                                     requests comments.
Number of tests for defects   0 to 5 tests for      Assumption based on
 per ingredient lot.           defects.              discussions with
                                                     industry--FDA
                                                     request comments.
Number of unlisted            0 to 6 components; 4  Ref. E47.
 components.                   most likely.
Number of tests per unlisted  1 identity test per   Assumption based on
 components.                   component.            discusssions with
                                                     industry--FDA
                                                     requests comments.
Number of shipments (Lots)    1 to 12 batches per   Assumption based on
 of ingredients and unlisted   shipment lot of       discussions with
 components.                   dietary ingredients.  industry--FDA
                                                     requests comments
                                                     (Ref. E48).
Number of batches produced..  Very small            Ref. E2.
                               establishments--223
                               ; Small
                               establishments--554
                               ; Large--309.
Number of inprocess           0 to 5 potential      Assumption based on
 potential defects.            control points; 2.5   discussions with
                               average.              industry--FDA
                                                     requests comments.
Number of inprocess tests     1 test per defect     Assumption based on
 per control point.            per control point.    discussions with
                                                     industry--FDA
                                                     requests comments.
Number of ingredients         Vitamins and          Assumption based on
 identified per identity       minerals--1 to 30;    discussions with
 test.                         2 most likely; All    industry--FDA
                               other categories--1   requests comments.
                               to 2.
Number of final product       3 tests batch.......  Assumption based on
 tests per batch.                                    discussions with
                                                     industry--FDA
                                                     requests comments.
Costs per test..............  Beta per              Refs. E50 and E51.
                               distribution skewed
                               rightward between
                               $20 to $150; $50
                               most likely; $60
                               average.
------------------------------------------------------------------------

    vi. Labor costs. We used the average manufacturing wage of $15.65 
per hour to estimate the cost of labor. We assumed that various tasks 
required by the proposed rule would take some number of hours per year, 
per batch of product, or per square foot of physical plant. For 
example, we assumed that time spent on the sanitation of physical 
plants is a function of the square footage. We assumed 1 hour per week 
for very small establishments, 3 hours per week for small 
establishments, and 20 hours per week for large establishments. We 
request comment or data about costs, hours, and the other requirements 
for these proposed required procedures.
    vii. Other costs. The main costs in this category are for pest and 
rodent control. We consulted a commercial supplier of these services 
for the estimated monthly costs, which were $400 to $600 a month for 
very small establishments, $480 to $720 for small establishments, and 
$700 to $1,000 for large establishments (Ref. E52). For each size of 
establishment, we selected the midpoint of the range as the most likely 
value.
    d. Estimating costs. We initially gathered information and made 
assumptions about the full cost of a provision. We then adjusted these 
estimates to account for the many activities already being carried out, 
as well other activities that would not have to be carried out by all 
establishments. We used the survey to estimate the likelihood that an 
establishment would incur a cost. To get an estimate of the average 
cost of provision (adjusted for baseline activities) for each category, 
we multiplied the average cost per establishment by the probability 
that the establishment would need to undertake the expense (one minus 
the probability that the establishment was already doing it). For each 
provision of the proposed rule, the simulation carried out the 
following calculation:
    Cost per unit of analysis for each provision = number of units of 
analysis per establishment x probability that establishment incurs cost 
x adjustment for requirement (yes or no) = cost per provision per 
establishment

    We estimated both a setup cost (a one-time fixed cost) of the 
provision and an annual recurring cost. The first-year costs would be 
the setup costs plus the annual costs. To get the total costs of the 
rule, we multiplied the number of establishments in each size category 
(from the survey) by the average costs per establishment in that 
category. We then adjusted for the establishments that did not respond 
to the survey but are believed to be in the industry. Two hundred 
thirty eight establishments responded to the survey; we estimated that 
1,566 firms are in the industry. We estimated costs with the following 
calculation:

[Number of very small establishments x costs per very small 
establishment) + (Number of small establishments x costs per small 
establishment) + (number of large establishments x costs per large 
establishment)]
x adjustment for establishments not in survey

The rule is complex and the industry is made up of very different kinds 
of firms, so cost estimates are averages with, in some cases, large 
variances. The cost per unit, number of batches and employees, and 
probability that the establishment would incur the cost all contain 
uncertainty. The values in table 15 of this document are used in the 
cost estimates, and are generated from multiple sources.

               Table 15.--Values Used in Cost Calculations
------------------------------------------------------------------------
                                    Value or
            Name                distribution used          Source
------------------------------------------------------------------------
Average wage per hour.......  $15.65..............  Employment Index,
                                                     Bureau of Labor
                                                     Statistics.
Average size of               Very small = 24,674;  Ref. E2.
 establishments in square      small = 71,354;
 feet.                         large = 596,000.
Average number of employees.  Very small = 7.6;     Ref. E2.
                               small = 95; large =
                               1,005.
Average annual number of      Very small = 223;     Ref. E44.
 batches.                      small = 554; large
                               = 309.

[[Page 12242]]

Annual time recordkeeping...  1/10 of setup time    Ref. E44.
                               per provision.
Personnel sanitation........  1 hour per week per   Assumption, based on
                               worker.               requirements of
                                                     proposed rule.
Sanitation time for physical  1 hour per week for   Assumption, based on
 plant.                        very small            difference in
                               establishments; 3     average physical
                               hours per week for    plant size.
                               small
                               establishments; 20
                               hours for physical
                               plant per week for
                               large
                               establishments.
Sanitation supervisor.......  Very small and small  Assumption, based on
                               establishments = 1    number of workers.
                               hour per week;
                               large
                               establishments = 1
                               hour per week.
Pest control setup costs....  $1,500 to $2,000 for  Ref. E52.
                               very small
                               establishments;
                               $1,800 to $2,400
                               for small
                               establishments;
                               $2,600 to $3,400
                               for large
                               establishments.
                               Average for each
                               size establishment
                               was midpoint
                               ($1,750, $2,100,
                               $3,000).
Pest control annual costs...  $400 to $600 per      Ref. E52.
                               month for very
                               small
                               establishments;
                               $480 to $720 for
                               small
                               establishments;
                               $700 to $1,000 for
                               large
                               establishments.
                               Average for each
                               size establishment
                               was the midpoint
                               ($500, $600, $850).
Renovation cost.............  $50 per square foot;  Based on
                               with 0 to 20          construction costs
                               percent of physical   and square feet.
                               plant to be
                               renovated, with 10
                               percent most likely.
Minimum quality control unit  1 person or 1         Assumption based on
                               percent of            requirements of
                               establishment work    proposed rule.
                               force.
Equipment replacement.......  For very small        Assumption, based on
                               establishments, 0     size of
                               to $1,000, with       establishments.
                               $100 most likely;
                               small, 0 to
                               $10,000, with
                               $1,000 most likely;
                               large, 0 to
                               $100,000 with
                               $1,000 most likely.
Setup costs for automatic     $500 for hardware,    Software costs and
 equipment.                    16 hours.             assumptions about
                                                     labor hours.
Annual costs for automatic    1 to 2 hours per      Assumption based on
 equipment.                    month for very        average size of
                               small and small       establishments.
                               establishments; 2
                               to 4 hours per
                               month for large
                               establishments.
Sanitation of equipment and   5 hours per week for  Assumption based on
 surfaces.                     very small            average sizes of
                               establishments, 15    establishments.
                               hours per week for
                               small
                               establishments, 100
                               hours per week for
                               large
                               establishments.
Number of dietary             12.8; standard        Ref. E46.
 ingredients per batch,        deviation = 15.6.
 supplements other than
 vitamins.
Number of dietary             3.6; standard         Ref. E46.
 ingredients per batch,        deviation = 4.8.
 supplements other than
 vitamins.
Cost per test...............  $20 to $150, with     See text discussion.
                               $50 most likely.
Holding products and dietary  Setup cost for very   Based on average
 ingredients: capital          small 0 to $1,000,    sizes of
 requirements.                 with $100 most        establishments.
                               likely. Multiply by
                               3 for small
                               establishments and
                               by 20 for large
                               establishments.
Default probabilities that    For very small        Based on results of
 establishments are not        establishments,       survey for other
 currently acting in           0.2; for small        practices.
 accordance with a provision.  establishments,
                               0.1, for large
                               establishments,
                               0.01.
------------------------------------------------------------------------

    We combined the costs per establishment with the number of 
establishments and probabilities from the survey, and adjusted for 
establishments not in the survey to estimate the total costs of the 
proposed rule. Table 16 of this document summarizes the estimated total 
costs for very small establishments, small establishments, large 
establishments, and warehouses. Table 17 of this document shows the 
total costs for the first year and annually after the first year, 
assuming that the proposed rule is phased in over 3 years. Table 18 of 
this document shows the total costs of the proposed rule compared to 
the total costs of other options.

                              Table 16.--Summary of Costs by Size of Establishment
----------------------------------------------------------------------------------------------------------------
                                                                                         Total 1st      Total
                                           Number of    1st Year costs   Annual costs    year costs     annual
                                        establishments        per             per           (in       costs  (in
                                                        establishments  establishments   millions)    millions)
----------------------------------------------------------------------------------------------------------------
Very small establishments.............            740         $62,000         $38,000           $46          $28
Small establishments..................            766          99,000          61,000            76           47
Large establishments..................             60          83,000          47,000             5            3
Warehouses and other holders..........         26,617             436             342            12            9
----------------------------------------------------------------------------------------------------------------

[[Page 12243]]

                                        Table 17.--Estimated Total Costs
                                                  [In millions]
----------------------------------------------------------------------------------------------------------------
                                                                                                       4th Year
                                                                1st Year     2nd Year     3rd Year    and after
----------------------------------------------------------------------------------------------------------------
Very small establishments...................................            0            0          $46          $28
Small establishments........................................            0          $76           47           47
Large establishments........................................            5            3            3            3
Warehouses..................................................           12            9            9            9
                                                             --------------
    Total...................................................           17           88          105           87
----------------------------------------------------------------------------------------------------------------

8. Summary of Benefits and Costs
    We estimated that, once it is fully implemented, the measured 
annual benefits from the proposed rule would be $218 million; measured 
annual costs would be about $86 million. Additional but unmeasured 
benefits should also be recognized when comparing the total costs and 
benefits. Table 18 of this document compares the benefits and costs of 
the proposed rule to the benefits and costs of the leading regulatory 
options. Because the phase in period, complicates the comparison for 
the early years, we limit the comparison to annual benefits once all 
establishments are covered.

        Table 18.--Annual Benfits and Costs of Regulatory Options
                              [In millions]
------------------------------------------------------------------------
                                                   Annual       Annual
               Regulatory option                  benefits      costs
------------------------------------------------------------------------
Proposed rule.................................         $218          $86
Fewer requirements for vitamins and minerals..          109           69
Stricter CGMP.................................          218          178
HACCP only....................................           42           38
Testing only (unable to estimate).............  ...........           32
High risk products only (unable to estimate)..  ...........       (\1\)
------------------------------------------------------------------------
\1\ Less than $86 million.

    Uncertainties in the analysis. In this section, we list all of the 
significant assumptions in the analysis, which if varied, could 
significantly change the estimates of costs and benefits. Such changes 
could have importance for the construction of any potential final rule. 
Therefore, we ask that comments address these aspects of the analysis 
and, where possible, provide FDA with better data to reduce the 
uncertainty. We estimated the benefits using indirect methods, which 
required several key assumptions that are critical for our estimates. 
With the exception of the recall benefit, which is based directly on 
FDA recall records, each component of the estimated benefits involves 
assumptions that reflect our uncertainty.
    Our basic assumption is that manufacturers lack market-based 
incentives to prevent hidden product quality defects. Our survey (Ref. 
E2) indicated that many firms do not have reliable quality control 
mechanisms in place. The survey was a one-time look at the 
manufacturing practices during the time of the survey. If the trend in 
the market is toward the adoption of the controls that we are proposing 
here in the absence of regulation, then both the cost and benefits of 
the rule will be less than we estimate. If the market-based trend is 
toward fewer controls, then both the cost and benefits of the 
regulation will be greater. Other key assumptions are listed below:
    The assumptions for the health benefits from reducing the number of 
sporadic illnesses model are:
    1. The baseline health of consumers is normal, not perfect. To 
estimate the change in health status from consuming defective products, 
we assumed that the baseline health of consumers is normal, which does 
not mean that we assumed that consumers have perfect health. We 
recognize that consumers will already have ``background'' health 
problems, by which we mean that many will have health problems 
unrelated to the consumption of defective products. Our assumption is 
that only the change in health status is relevant for our analysis. If 
an immune-compromised consumer is made ill by a defective product, 
e.g., gets lead poisoning, the consumer might in fact have more 
difficulty recovering than an otherwise healthy person. However, we 
assume that the change in productivity, functional state, pain and 
suffering, and medical costs will be the same, regardless of prior 
health status. Accounting for confounding factors would have the effect 
of making health problems worse than we estimate, not better, so our 
estimate may be understating the true health benefits.
    2. The average value of a QALY is $630 per day. That value, $630 
per day, is in turn based on: (1) The value of a statistical life of $5 
million; (2) the expected remaining life of consumers of 21.84 years 
(average), discounted from 36 years; and, (3) the social rate of time 
preference of 3 percent. The estimate is derived from workers in 
somewhat risky occupations who demand a wage premium for their 
additional risk of fatality. If our estimate of the value of a 
statistical life of workers does not represent the value of a 
statistical life of consumers of dietary supplements, then our benefits 
estimate will be different from the true health benefits of the rule. 
If consumers value their life differently than workers or if consumers 
place different values for different kinds of hazard-related deaths 
than do workers for job-related safety hazards, then we will have 
incorrect estimates for the true health benefits. If we discount life 
expectancy by 7 percent instead of 3 percent, the benefits would be 
much higher.
    3. There is one illness for each recall. We assumed that for each 
class 1 and 2 recalled product there was only one illness that was 
reported to the public health authority. For instance, if a product was 
recalled because the defective product contained lead, we assume that a 
person was made ill from lead poisoning and that was how the recalled 
product was discovered. If there were more illnesses per recall than 
one, then our estimates of benefits will be low. If fewer than one 
illness per recall occurred (or is likely to occur in the future), then 
our estimate of health benefits will be more than the actual health 
benefits.
    4. The assumed frequency of actual illnesses is 100 times the 
frequency of reported illnesses. This assumption is based on Ref. E16. 
We recognize that the factor of 100, although it has empirical support, 
might be wrong and that there is likely to be considerable uncertainty 
about this point estimate. It is widely believed in the public health 
community that most illnesses are underreported to public health 
authorities, particularly in passive reporting systems, such as the 
case with

[[Page 12244]]

dietary supplements. Mild cases are the most underreported. For 
instance, victims rarely notify public health authorities when they 
have minor gastrointestinal tract related illnesses. It is even more 
rare to report the likely source of a mild illness. It is also widely 
believed that severe illnesses and death are reported much more 
frequently than milder illnesses, even when the cause of illness or 
death is not included in the report. Although the number of deaths that 
are reported probably approach 100 percent, the cause of death from a 
contaminated dietary supplement product might not be reported. We 
believe that using a single composite factor--100--to represent the 
total number of all unreported cases, including mild, severe, and 
death, does not invalidate our assumption. The factor of 100 represents 
an estimate of the composite probability of the full range of 
probabilities for each severity level of an illness being reported. 
Increasing the factor multiplier from 100 to some number higher would 
increase the health benefits, while lowering the multiplier would 
decrease the health benefits. If we assume that all illnesses are 
reported--there are no unreported illnesses and no factor of 100, then 
the health benefits from fewer sporadic illnesses will be less than $1 
million.
    5. Introducing CGMP's will reduce the probability of a recall to 
zero. We believe that the proposed CGMP's creates the most reliable 
means for discovering product adulteration. Indeed, we believe that it 
will, if strictly used, cause the discovery of all adulteration. 
Therefore, we assume that once an establishment fully adopts the 
requirements, there should be no more health risk from adulterated 
dietary supplements and consequently, no more class 1 and 2 recalls. 
This conclusion rests on the assumption that there will be 100 percent 
compliance with this regulation. We recognize that human error is 
inescapable. If recalls--or a health risk from adulteration--would 
still exist, then we overstated the true health benefits of the 
regulation.
    The assumptions for the health benefits from lowering the 
likelihood of rare catastrophic event model are:
    1. We assume that a rare catastrophic event would occur every 30 
years. We recognize that the occurrence of a single event provides 
little evidence about what will happen in the future. If the event 
reported in this analysis was in fact a one-time occurrence, then our 
estimate of the benefits from the prevention of the catastrophic health 
event would overstate the true benefits, which in fact should be zero. 
There would have been no future event, and there would be no benefit 
from adopting a rule to avoid it. If a rare event would have happened 
more frequently than our estimate of once every 30 years, then our 
estimate of the benefits would underestimate the true health benefits.
    2. Number of illnesses per rare event. We based our estimate of the 
health impact from contaminated L-Tryptophan. If the number of 
illnesses from a future rare event differed--either more or less--then 
the health benefits would differ from our estimated benefits. If a 
future event would have had 10,000 cases, not 1,500 cases, then our 
estimate would understate the true health benefits of avoiding such a 
large catastrophe.
    The assumptions for fewer products recalled are:
    1. The reported class 1 and 2 recalls that have occurred over the 
last 10 years represent the number and type of recalls that would have 
occurred in the future but for the implementation of this regulation. 
If the number or types of recalls are not representative, then we over 
or under estimated the benefit of avoiding recalls. Avoiding one very 
large recall could result in significantly higher benefits. Conversely, 
merely avoiding fewer or smaller recalls would result in smaller 
benefits.
    2. A product recall causes sellers to lose both goodwill and the 
value of the recalled product and lost goodwill equals the value of the 
recalled product. These two embedded assumptions have empirical support 
from Ref. E24. A product recall adversely affects the wealth of 
sellers--a recall leads to lost goodwill--by signaling to consumers 
that products are defective. From evaluating the declines in public 
share prices after product recalls in various industries, the authors 
in Ref. E24 determined that the loss in share price is twice the value 
of the loss of the actual value of the product recalled. They attribute 
the difference to lost firm goodwill.
    3. Full compliance with the proposed CGMP's will reduce the 
probability of a recall to zero. As in our earlier assumption about the 
probability of recalls after the rule is adopted, consistency requires 
that if we believe that the rule will reliably cause the discovery of 
adulterated products before they are commercially available, there 
should be no more health risk from adulterated dietary supplements. 
Consequently, there should be no more recalls.
    We developed the hypothetical search model to estimate the implicit 
value to consumers of better product quality although we lacked a model 
that could enable us to directly estimate consumer preferences for 
dietary supplement quality. With the adoption of the proposed rule, the 
standardization of manufacturing practices will reduce product 
differentiation. In a perfect information market, the change in product 
differentiation would be reflected in the change in the price 
differences between low and high quality products. In the existing 
market, price differences alone are an inadequate signal because the 
differences in product quality are typically hidden from the view of 
both consumers and (though less so) manufacturers. In this hypothetical 
model, we assumed that if there were actually indicators of product 
quality in the market now, consumers would spend a certain amount of 
time attempting to find a reasonably high quality product. Time spent 
searching is an economic cost. In fact, in markets where quality is 
discernible prior to purchase, such search does take place and it is 
from those markets that our estimates were derived. In such a world of 
easily available product quality signals, this regulation, by 
standardizing product quality at the high end, would reduce that search 
time. Our assumption is that this is a reasonable indicator of 
consumers' value for high quality products. Further, we assume that in 
fact consumers of dietary supplements do wish to purchase high quality 
products, as the absence of quality could mean either an ineffective 
product or worse, illness or death. We used various assumptions at each 
step in our model, and the benefits change when the assumptions change. 
The assumptions that we used for the search model are:
    1. Consumers will search until the expected benefits of the search 
equal the expected cost of additional search. The expected cost is the 
value of their time, which we estimated is the average wage rate for 
manufacturing workers--$15.65/hour. If the true wage rate is different, 
the benefits of the rule will be different.
    2. The three models--drug store, use of time and grocery store 
models--represent consumers of dietary supplements. If not, then we 
will not have estimated the true preferences of consumers. If consumers 
value dietary supplements more highly than either drugs, groceries or 
other uses of time, and they search more for better quality, then we 
understated the benefits of product standardization. If consumers value 
dietary supplements less highly than either drugs, they search more for 
better quality, then we overstated the benefits.
    3. The quality controls will reduce consumer search time by 
approximately

[[Page 12245]]

33 percent. If our estimate is not representative of the true average 
reduction, then our estimate will be wrong.
    4. The type and number of consumers represent the true value. If 
children, the elderly or other consumers search for these products in 
significantly greater amounts than average workers or the estimated 
population, then we may have overstated the benefits, because their 
foregone wages would be less than that of average workers.
    In an ideal analysis, the benefits and costs of each provision 
would be evaluated. We were not able to quantify the benefits for each 
of the provisions in our analysis although we do have fairly detailed 
estimates of the cost. We request comments on marginal costs and 
benefits of specific provisions in the rule. Comments can be directed 
either at how well a specific provision might work to make dietary 
supplements either safer or of higher quality, or be directed at the 
cost of the provision. An example of this type of provision follows for 
recordkeeping:
Benefits of Recordkeeping
    Mandatory recordkeeping is intended to help the discovery of 
manufacturing practices that create defective products. Recordkeeping 
ensures that preventative controls are carried out for each batch of 
dietary supplements produced. Records serve as a checklist that quality 
control personnel can consult to monitor that necessary controls are 
implemented or corrective actions taken. Further, mandatory 
recordkeeping provides an incentive for manufacturers to comply more 
fully with the provisions of the rule where recordkeeping is required. 
Knowing that FDA inspectors will examine records and that falsifying 
them is a criminal offense provides strong incentives to keep thorough 
and accurate records that the required safety functions have been 
performed adequately and in a timely manner. Thus, the benefits of 
recordkeeping are to permit detection of defective products and 
increase compliance with the provisions for which recordkeeping is 
required. If, for example: (1) The total benefits of the requirements 
that have recordkeeping attached to them were $50 million (not the real 
value); (2) only half of the requirements would be met without 
recordkeeping; and, (3) recordkeeping raised the compliance rate to 100 
percent, then the benefits of recordkeeping would be $25 million. We 
were not able to quantify the marginal benefits of this requirement 
with numbers like this. Comments are requested for how well records are 
likely to perform this function. We estimate that the additional cost 
to society for the proposed new recordkeeping requirement will be 
approximately 10 percent of the total annual cost of the proposed 
regulation, or a little less than $9 million per year.
    Further, we request comments on all of the provisions that would be 
of a similar nature to this example.
    The costs of the rule depend on our assumptions about the amount 
and cost of testing. The amount of testing is highly uncertain; we have 
tried to model the number of tests based on number of ingredients and 
types of tests.
    We first characterized the uncertainty as a probability 
distribution. We ran 1,000 computer simulations to estimate both 
benefits and costs. The simulations used distributions and assumptions 
from tables 8 through 13 of this document in place of single estimates.

                   Table 19.--Distribution of Simulation Results for Annual Benefits and Costs
                                                  [In millions]
----------------------------------------------------------------------------------------------------------------
                                                                  5th                                    95th
                                                               Percentile     Median        Mean      Percentile
----------------------------------------------------------------------------------------------------------------
Annual benefits.............................................          $89         $198         $218         $405
Annual costs................................................           62           80           86          128
----------------------------------------------------------------------------------------------------------------

    The computer simulation gives the distribution of estimated 
benefits and costs. If the underlying distributions capture the 
uncertainty of the estimates, then the results in table 19 of this 
document give a clear picture of the uncertainty. Another way to show 
the uncertainty is to see how sensitive the results are to plausible 
changes in individual variables. We start with benefits.

                   Table 20.--Sensitivity of Benefits
                              [In millions]
------------------------------------------------------------------------
                                                             Estimated
                       Description                            annual
                                                             benefits
------------------------------------------------------------------------
The proposed rule.......................................            $218
If reporting rate of illness is 0.1 (baseline is 0.01)..             182
If reporting rate of illness is 0.005 (baseline is 0.01)             257
If the value of a statistical life is $3 million                     175
 (baseline is $5 million)...............................
If the value of a statistical life is $7 million                     259
 (baseline is $5 million)...............................
If consumer search time per item is 1 minute (baseline               137
 is 3.75 minutes).......................................
If consumer search time per item is 5 minutes (baseline              250
 is 3.75 minutes).......................................
If consumer search time equals 40 percent of shopping                166
 time (baseline is 70 percent)..........................
If consumer search time is equal to shopping time                    254
 (baseline is 70 percent)...............................
If consumer search for quality accounts for 30 percent               278
 of search time (baseline is 20 percent)................
If consumer search time for quality accounts for 10                  158
 percent of search time (baseline is 20 percent)........
If catastrophic events are not prevented (baseline is                152
 $66 million annual benefit from prevention)............
------------------------------------------------------------------------

    We mainly looked at the cost effects of changing assumptions about 
testing and consumer complaints. As table 21 of this document shows, 
annual costs are quite sensitive to the assumptions about the average 
cost and number of tests.

[[Page 12246]]

                     Table 21.--Sensitivity of Costs
                              [In millions]
------------------------------------------------------------------------
                                                             Estimated
                       Description                         Annual Costs
------------------------------------------------------------------------
The proposed rule.......................................             $86
6 tests per batch (baseline is 3).......................             119
1 test per batch (baseline is 3)........................              66
$100 per test (baseline is $60).........................             101
1 consumer complaint per 20 batches (baseline is 1 per                77
 10)....................................................
1 consumer complaint per 5 batches (baseline is 1 per                104
 10)....................................................
------------------------------------------------------------------------

C. Initial Regulatory Flexibility Analysis

1. Introduction
    FDA has examined the economic implications of this proposed rule as 
required by the Regulatory Flexibility Act (5 U.S.C. 601-612). If a 
rule has a significant economic impact on a substantial number of small 
entities, the Regulatory Flexibility Act requires agencies to analyze 
regulatory options that would lessen the economic effect of the rule on 
small entities. We find that this proposed rule would have a 
significant economic impact on a substantial number of small entities.
2. Economic Effects on Small Entities
    a. Number of small entities affected. The proposed regulations 
would affect many small entities. Our classification of establishment 
size is based on the Small Business Administration's definition for 
small, as discussed previously in this document. A small business by 
this definition is any establishment with fewer than 500 employees. For 
this analysis, we defined very small establishments as establishments 
with fewer than 20 employees. Some small and very small establishments 
produce very large revenues and would probably not incur a large 
decline in profitability from the proposed CGMP regulations. We lack 
precise information about those establishments. Based on the survey, we 
estimated that 830 establishments, 53 percent of the total 
establishments, could be classified as very small (under 20 employees) 
and 564 as small (20 to 499 employees), which is 36 percent of the 
total establishments.
    We estimated that 95 percent of all holders (warehouses and 
wholesalers) covered by this regulation are small using the Small 
Business Administration definition. The total number of holders likely 
to be affected by this regulation is 26,617 (see table 4 of this 
document), so the total number of holders that are small would be 
25,286 (= 0.95 x 26,617).
    The small establishments that would be affected by the proposed 
regulations are those establishments that would have to perform the 
various required activities, and that would not have done so without 
the regulations. As in the preliminary regulatory impact analysis 
(section VII.B of this document), we determined our estimate of 
baseline (pre-CGMP) manufacturing practices with the survey of the 
industry (Ref. E2). The survey asked representative respondents to 
answer a series of questions, including how many employees they had and 
what their existing practices were. From the survey, we determined that 
small establishments do not now follow all of the provisions of the 
proposed CGMP regulations now. Those that do not follow the proposed 
requirements will incur a cost to do so.
    b. Costs to small entities. Implementation costs vary across 
establishments based on current practices and the types of products 
manufactured, packaged, or held. We estimated the range of current 
practices using the survey of the industry. The cost model divided 
establishments by size, which allowed us to estimate the distribution 
of costs per establishment for each size and product class. Table 22 of 
this document shows the cost per establishment for very small and small 
establishments. For comparison, we include the estimated average cost 
per large establishment and the median revenues for each size category. 
As the table shows, costs per establishment are proportionally higher 
for very small than for large establishments. The table's most striking 
result is that costs are highest for small (20 to 499 employees) 
establishments.

                    Table 22.--Cost Per Establishment
------------------------------------------------------------------------
                                             1st year         Annual
------------------------------------------------------------------------
Very small--fewer than 20 employees;             $62,000         $38,000
 median revenue under $1 million........
Small--20 to 499 employees; median                99,000          61,000
 revenue $5 to 10 million...............
Large--500 or more employees; median              83,000          47,000
 revenue $20 to $50 million.............
------------------------------------------------------------------------

    Small establishments that do not perform a substantial number of 
the actions required by the proposed CGMP regulations would bear 
relatively high costs for compliance with the provisions of this 
proposed rule. As shown in table 22 of this document, we estimated the 
average annual compliance costs for a very small establishment to be 
around $38,000. About one-third of those establishments or about 500 
firms have annual sales revenues under $500,000. In addition, the 
average annual compliance cost for a small establishment is around 
$61,000. As the survey indicated, about 14 percent of establishments 
with 20 to 499 employees or about 200 firms have annual sales revenues 
under $500,000. For purposes of our analysis, we regard firms with 
revenues of $500,000 or less to be low revenue firms. Although the 
proposed rule would raise product prices, the price increase (which 
would largely be determined by changes made by large establishments) 
would be much smaller than the increase in the average costs of very 
small producers. The average burden to very small low revenue firms, 
then, would be at least 8 percent of their annual revenue. The average 
burden to small low revenue firms would be at least 12 percent of 
annual revenue. Establishments with above average costs, and even 
establishments with average costs,

[[Page 12247]]

would be hard pressed to continue to operate. Therefore, some of these 
establishments, for example, such as those that produce other products 
(foods or pharmaceuticals) or are part of firms with more than one 
establishment, may decide it is too costly and either change product 
lines or go out of business. If we assume that one half of these firms 
have sales revenues from other products and locations and remove them 
from the at-risk group, we are left with approximately 350 very small 
and small establishments with less than $500,000 in revenue. It is 
possible that a large number of these 350 very small and small 
establishments would be unable to absorb the compliance costs and will 
close.
3. Regulatory Options
    a. Exemptions for small entities. The burden on small 
establishments would be reduced if they were exempt from some 
provisions of the proposed rule. Most entities affected by this 
proposed rule, however, are small. Exempting small establishments from 
some or all of its provisions would be likely to reduce benefits.
    b. Longer compliance periods. Lengthening the compliance period 
would provide regulatory relief for small entities. A longer compliance 
period for small entities would allow additional time for setting up 
recordkeeping, making capital improvements to the physical plant, 
purchasing new or replacement equipment, and other one-time 
expenditures. It would also delay the impact of the annual costs of 
compliance. We have given very small and small firms an additional 2 
years for compliance. The proposed rule, then, would be phased-in over 
3 years, with large firms complying after 1 year, and both very small 
and small firms after 3 years. After 3 years, the annual costs would be 
incurred. The cost savings of delay may well be larger than simply the 
present value of the delay because very small and small firms may also 
be able to reduce their compliance costs by taking advantage of 
increases in industry knowledge and experience in implementing CGMP 
regulations. A summary of the compliance costs is shown in table 22 of 
this document.
    Although lengthening the compliance period would provide some 
regulatory relief to small entities, relief for these provisions would 
also delay the full realization of the benefits of the proposed rule.
4. Description of Recordkeeping and Reporting
    The Regulatory Flexibility Act requires a description of the 
recordkeeping and recording required for compliance with this proposed 
rule. This proposed rule would require the preparation of records. As 
described in the Preliminary Regulatory Impact Analysis, records must 
be written or electronic documents must be kept that demonstrate that 
specific action or actions occurred in the manufacturing process in 
compliance with the proposed regulations. Records that would be 
required in this proposed rule would demonstrate, that corrective 
actions were taken, that equipment, instruments, and controls used in 
laboratory operations and quality control were installed properly, and 
calibrated; that maintenance programs were followed; and that the 
results of any testing meet the necessary specifications.
    The compliance cost of recordkeeping is the sum of both the initial 
design and printing of the recordkeeping documents and the recurring 
costs of maintaining the records. The cost of training personnel to use 
the new documents is a recurring cost depending on how frequently 
documents are modified, how often personnel turn over, and how 
complicated the tasks are that are being recorded. The recurring costs 
are measured by the workers' wage rate multiplied by the expected labor 
hours necessary to perform a written or electronic record and the time 
necessary for management to review the records to see that actions are 
documented accurately. In addition, electronic records necessitate 
recurring time spent ensuring that the equipment is serviced and 
maintained properly.
5. Summary
    The proposed CGMP regulations would have a significant economic 
impact on a substantial number of small entities.

D. Unfunded Mandates

    The Unfunded Mandates Reform Act of 1995 (Pub. L. 104-4) requires 
cost-benefit and other analyses for rules that would cost more than 
$100 million in a single year. The current inflation-adjusted statutory 
threshold is $112 million. The proposed rule qualifies as a significant 
rule under the statute because there is a significant possibility that 
the cost of the rule will be above the threshold. Most of the 
requirements of the Unfunded Mandates are fulfilled in the Executive 
Order 12866 analysis. The requirements under the Unfunded Mandates Act 
of 1995 include assessing the rule's effects on future costs; 
productivity; particular regions, communities, or industrial sectors; 
economic growth; full employment; job creation; and exports.
Future Costs
    The future costs from the rule include the recurring costs, which 
reach their long-term value in the third year after the proposed rule 
would become final. These costs would be incurred by the establishments 
that manufacture, process, pack, transport, distribute, receive, hold, 
or import dietary ingredients or dietary products. Recurring costs from 
the regulatory requirements would be incurred in each future year. 
Table 18 of this document summarizes the annual future recurring costs.
Particular Regions, Communities, or Industrial Sectors
    The costs of the rule will be shared among manufacturers, 
processors, packagers, transporters, receivers, holders, and importers 
of dietary ingredients or dietary products as well as domestic 
consumers. The higher costs incurred by domestic suppliers of dietary 
supplement products as a result of these regulations will mostly be 
passed on to consumers in the form of higher prices. Since consumer 
demand for dietary supplements is price elastic, most of the higher 
costs incurred by suppliers will be passed on to consumers. 
Consequently, higher dietary supplement prices will reduce real incomes 
for many consumers. However, the reduction in real incomes is thought 
to be more than offset by the benefits from these regulations. These 
benefits are measured as an improved ability by the FDA to respond to 
and contain threats of serious adverse health consequences from 
accidental contamination of dietary supplements.
National Productivity, Economic Growth, Job Creation, and Full 
Employment
    Although this proposed regulation is significant, we do not expect 
it to substantially affect national productivity, growth, jobs, or full 
employment. The total costs will be small relative to the economy, and 
will be offset by benefits. The improved ability to respond to, and 
contain, serious adverse health consequences means less illness and 
fewer sick days taken by employees, and lower adjustment costs by firms 
that would otherwise need to hire replacement employees.

[[Page 12248]]

Exports
    This proposed rule would require additional controls to be kept 
throughout the production and distribution chain for the manufacture of 
dietary ingredients and dietary supplements. The additional control 
costs would increase the total costs of production and distribution for 
all of the regulated products, including products sold within the U.S. 
and across national borders. These increased costs will be largely 
passed on to consumers in the form of higher prices, which will tend to 
reduce the quantity demanded of the regulated products. The increased 
prices of U.S. exports could reduce the quantity of U.S. exports 
demanded, particularly in comparison with exports from countries that 
do not implement similar regulations. We expect this effect to be 
insignificant, because under the proposed rule the increases in the 
price of United States exports (and resulting decreases in quantity 
demanded) would be quite small.

VIII. Federalism

    FDA has analyzed this proposed rule in accordance with the 
principles set forth in Executive Order 13132. We consulted with seven 
State officials to make a tentative determination about whether this 
proposed rule would have federalism implications. Based on this 
consultation, it does not appear that this proposed rule has federalism 
implications. In addition, we sent a letter on March 7, 2000, to 
elected State officials and their representative organization to notify 
them that our unified agenda was published on November 22, 1999, and 
identified this proposed CGMP rule as a rule that would publish in the 
year 2000. In that letter, we solicited comments on any federalism 
implications that this proposed rule may have. To date, no responses 
have been received to our solicitation. After publishing this proposed 
rule, FDA will send a letter to elected State officials and their 
representative organization requesting consultation about any 
federalism implications. We invite comment on our tentative 
determination that this proposed rule does not have federalism 
implications, and therefore, does not contain policies that have 
substantial direct effects on the States, or on the distribution of 
power and responsibilities among the various levels of government.

IX. Request for Comments

    Interested persons may submit to the Dockets Management Branch (see 
ADDRESSES) written or electronic comments regarding this document. 
Submit a single copy of electronic comments to http://www.fda.gov/
dockets/ecomments Exit Disclaimer
 or two hard copies of any written comments, except 
that individuals may submit one hard copy. Comments are to be 
identified with the docket number found in brackets in the heading of 
this document. Received comments may be seen in the Dockets Management 
Branch between 9 a.m. and 4 p.m., Monday through Friday.

X. References

    We have placed the following references on display in the Dockets 
Management Branch (see ADDRESSES). You may see them between 9 a.m. and 
4 p.m., Monday through Friday.
    1. Section 2, ``Findings,'' Dietary Supplement Health and Education 
Act, Pub. L. 103-417, pp. 4325-4326, October 25, 1994.
    2. ``Manufacturing Practices for Nutritional Supplements,'' United 
States Pharmacopeia, 12601 Twinbrook Pkwy., Rockville, MD 20852, 
General Chapter 2750, 2186-2192, and 2834, 1993.
    3. ``NNFA Good Manufacturing Practice in Manufacturing, Packing, or 
Holding Dietary Supplements,'' National Nutritional Foods Association, 
3931 MacArthur Blvd., suite 101, Newport Beach, CA 92660, 1999.
    4. ``Draft Report of the Food Advisory Committee Dietary Supplement 
Working Group on Ingredient Identity Testing and Records and 
Retention,'' FDA Food Advisory Committee Dietary Supplement Working 
Group, Center for Food Safety and Applied Nutrition, FDA, June 25, 
1999.
    5. ``Consumer Use of Dietary Supplements,'' Rodale Press, 733 Third 
Ave., New York, NY 10017, 2000.
    6. Slifman, N. R., W. R. Obermeyer, B. K. Aloi, S. M. Musser, W. A. 
Correll, S. M. Cichowicz, J. M. Betz, and L. A. Love, ``Contamination 
of Botanical Dietary Supplements by Digitalis Lanata,'' New England 
Journal of Medicine, 339:12, pp. 806-811, 1998.
    7. ``Cardiac Glycoside Drugs,'' In: Wagner, H., and S. Bladt, Plant 
Drug Analysis: A Thin Layer Chromatography Atlas, 2d ed., Berlin, 
Germany: Springer-Verlag, pp. 99-123, 1996.
    8. Bruneton, J., Pharmacognosy, Phytochemistry, Medicinal Plants, 
Paris, Lavoisier Publishing, pp. 572-598, 1995.
    9. Evans, W. C., Trease and Evans' Pharmacognosy. 14th ed., 
Philadelphia, W. B. Saunders, pp. 309-318, 1996.
    10. Moe, G. K. and A. E. Farah, ``Digitalis and Allied Cardiac 
Glycosides,'' In L. S. Goodman and A. Gilman, eds., The Pharmacological 
Basis of Therapeutics, 5th ed., New York. Macmillan Publishing, pp. 
653-682, 1975.
    11. ``Survey on Botanical Adulteration,'' American Herbal Products 
Association, 8484 Georgia Ave., suite 370, Silver Spring, MD 20910, 
March 4, 1998.
    12. Roy, A. K. and H. K. Chourasia, ``Mycotoxin Incidence in Root 
Drugs,'' International Journal of Crude Drug Residues, 28:2, pp. 157-
160, 1990.
    13. Kumar, S. and A. K. Roy, ``Occurrence of Aflatoxin in Some 
Liver Curative Herbal Medicines,'' Letters in Applied Microbiology, 17, 
pp. 112-114, 1993.
    14. Kumar, S. and G. Prasad, ``Efficacy of Medicinal Plant 
(Andrographis peniculata) Extract on Aflatoxin Production and Growth of 
Aspergillus flavus,'' Letters in Microbiology 15, pp. 131-132, 1992.
    15. ``Adverse Events Associated with Ingestion of Gamma-
Butyrolactone: Minnesota, New Mexico, and Texas, 1998-1999,'' MMWR 
Weekly, 48:07, pp. 137-140, February 26, 1999.
    16. FDA Establishment Inspection Report, Ducoa Co., CFN 
#1944182, August 31, 1998 and September 1, 1998.
    17. FDA Health Hazard Evaluation, Classification, and FDA 
Enforcement Report for Firm-Initiated Recall--Lead, U.S. FDA Recall 
#F-006-5 and #F-137/145-5, 1994.
    18. FDA Health Hazard Evaluation, Classification, and FDA 
Enforcement Report for Firm-Initiated Recall Salmonella, U.S. FDA 
Recall #F-164/171-8, 1988.
    19. FDA Health Hazard Evaluation, Classification, and FDA 
Enforcement Report for Firm-Initiated Recall--Klebsiella Pneumonia, 
U.S. FDA Recall #F-522-0, 1990.
    20. FDA Health Hazard Evaluation, Classification, and FDA 
Enforcement Report for Firm-Initiated Recall--Botulism, U.S. FDA Recall 
#F-563/564-9, 1989.
    21. FDA Health Hazard Evaluation, Classification, and FDA 
Enforcement Report for Firm-Initiated Recall--Glass Fragments, U.S. FDA 
Recall #F-540-3, 1993.
    22. FDA Health Hazard Evaluation, Classification, and FDA 
Enforcement Report for Firm-Initiated Recall Super-Potent Vitamin A, 
U.S. FDA Recall #F-157/158-5, 1994.
    23. FDA Health Hazard Evaluation, Classification, and FDA 
Enforcement Report for Firm-Initiated Recall--Super-Potent Vitamin D, 
U.S. FDA Recall #F0610-3, 1993.
    24. FDA Health Hazard Evaluation, Classification, and FDA 
Enforcement Report for Firm-Initiated Recall--Super-

[[Page 12249]]

Potent Vitamin B-6, U.S. FDA Recall #F-272-3, 1993.
    25. FDA Health Hazard Evaluation, Classification, and FDA 
Enforcement Report for Firm-Initiated Recall Super-Potent Selenium, 
U.S. FDA Recall #F-007-5, 1994.
    26. ``Table 1. Tolerable Upper Intake Levels (UL) for Certain 
Nutrients and Food Components,'' Dietary Reference Intakes: A Risk 
Assessment Model for Establishing Upper Intake Levels, 20, Food and 
Nutrition Board, Institute of Medicine, National Academy of Sciences, 
Washington, DC, June 1999.
    27. FDA Health Hazard Evaluation, Classification, and FDA 
Enforcement Report for Firm-Initiated Recall--Sub-Potent Folic Acid, 
U.S. FDA Recall #F-825-7, 1996.
    28. FDA Health Hazard Evaluation, Classification, and FDA 
Enforcement Report for Firm-Initiated Recall Color Additives, Yellow 
#6, Red #40, Blue #2, U.S. FDA Recall #F-314-8, 1998.
    29. FDA Health Hazard Evaluation, Classification, and FDA 
Enforcement Report for Firm-Initiated Recall--Color Additives, Yellow 
#5, U.S. FDA Recall #F-761-4, 1994.
    30. FDA Health Hazard Evaluation, Classification, and FDA 
Enforcement Report for Firm-Initiated Recall--Lactose, U.S. FDA Recall 
#F-271-6, 1995.
    31. FDA Health Hazard Evaluation, Classification, and FDA 
Enforcement Report for Firm-Initiated Recall--Sulfites, U.S. FDA Recall 
#F-538/539-9 1999.
    32. Weber, R. W., ``Food Additives and Allergy,'' Annals of 
Allergy, 70, pp. 183-190, 1993.
    33. Gurley, B. J., S. F. Gardner, and M. A. Hubbard, ``Content 
Versus Label Claims in Ephedra-containing Dietary Supplements,'' 
(Prepublication version) American Journal Health Systems Pharmacy, 57, 
pp. 963-969, May 15, 2000.
    34. Consumer Labs Product Reviews on Ginkgo Biloba, Saw Palmetto, 
Glucosamine and Chrondroitin, and SAMe, (http://www.consumerlabs.com).
    35. ``Methods Validation and Technical Programs,'' AOAC Official 
Methods Program, (http://www.aoac.org), Exit Disclaimer Exit Disclaimer March 17, 2000.
    36. Alexander, R. G., D. A. Wilson, and A. G. Davidson, ``Medicines 
Control Agency Investigation of the Microbial Quality of Herbal 
Products,'' The Pharmaceutical Journal, pp. 259-261, August 1997.
    37. Transcript, FDA and USDA Symposium, ``Tissue Distribution, 
Inactivation, and Transmission of Transmissible Spongiform 
Encephalopathies of Animals,'' Riverdale, MD, 1996.
    38. ``Frontier,'' ``Nature's Herbs,'' and ``Nature's Way,'' (http://
www.mothernature.com), Exit Disclaimer June 2000.
    39. USDA Food Composition Data, National Agricultural Library, Food 
and Nutrition Information Center, USDA, May 2000.
    40. ``King-size Gingersnaps, The Good Housekeeping Cookbook, p. 
564, New York, NY, 1973.
    41. ``Food Defect Action Levels: Levels of Natural or Unavoidable 
Defects in Foods that Present no Health Hazards for Humans,'' 
Department of Health and Human Services, Public Health Service, Food 
and Drug Administration, Center for Food Safety and Applied Nutrition, 
Washington, DC 20204, May 1998.
    42. ``Hazard Analysis and Critical Control Point Principles and 
Application Guidelines,'' National Advisory Committee on 
Microbiological Criteria for Foods, August 14, 1997.
    43. Presidential Memorandum on Plain Language, The White House, 
Washington, June 1, 1998.
    44. Olsen, A. R., ``Regulatory Action Criteria for Filth and Other 
Extraneous Materials. II. Allergenic Mites: An Emerging Food Safety 
Issue,'' Regulatory Toxicology and Pharmacology, 28, pp. 190-198, 1998.
    45. U.S. Department of Health and Human Services, Food and Drug 
Administration, ``Food Code,'' Section 1-201.10(B)(72), p. 15, 1999.
    46. ``Immunoglobin g (IgG) Dietary Supplement,'' 
(www.kingsnutrition.com/immunol.html).
    47. Vasarada, P. C. and M. A. Cousin, ``Dairy Microbiology and 
Safety,'' In: Dairy Science and Technology Handbook, vol. 2, edited by 
Y. H. Hui, New York, Chapter 5, pp. 301-426, VCH Publishers, Inc., 
1992.
    48. Stauffer, J. E., ``Quality Assurance and Dairy Processing,'' 
In: Dairy Science and Technology Handbook, vol. 3, edited by Y. H. Hui, 
New York: Chapter 1, pp. 1-76, VCH Publishers, Inc., 1992.
    49. Biotics Research Products (http://www.increasedhealth.com/
biotics_description.shtml). Exit Disclaimer
    50. ``Bovine-Derived Materials; Agency Letters to Manufacturers of 
FDA-Regulated Products,'' Federal Register, August 29, 1994.
    51. ``Points to Consider in the Manufacture and Testing of 
Therapeutic Products for Human Use Derived from Transgenic Animals,'' 
FDA, Center for Biologics Evaluation and Research, 1995.
    52. ``Guidance for Industry, The Sourcing and Processing of Gelatin 
to Reduce the Potential Risk Posed by Bovine Spongiform Encephalopathy 
(BSE) in FDA-Regulated Products for Human Use,'' U.S. Department of 
Health and Human Services, Food and Drug Administration, September 
1997.
    53. ``Guidance for Industry, Guide to Minimize Microbial Food 
Safety Hazards for Fresh Fruits and Vegetables,'' U.S. Department of 
Health and Human Services, Food and Drug Administration, Center for 
Food Safety and Applied Nutrition (CFSAN), October 1998.
    54. Webster's II New Riverside University Dictionary, Houghton 
Mifflin Co., 222 Berkeley St., Boston, MA 02116, p. 961, 1994.
    55. ``FDA Fact Sheet: Shigella in Food,'' Food and Drug 
Administration, U.S. Department of Health, Education, and Welfare, 
Public Health Service, December 1969.
    56. Bryan, F. L., ``Emerging Foodborne Diseases: I. Their 
Surveillance and Epidemiology,'' Journal of Milk and Food Technology, 
35:10, pp. 618-625, October 1972.
    57. Cross Connection Control Committee, Pacific Northwest Section, 
American Water Works Association, with the Assistance of EPA, ``Cross 
Connection Control Manual Accepted Procedures and Practice,'' Section 1 
``Introduction to Cross Connection Control and Backflow Prevention,'' 
5th ed., pp. xiv, 1-2, 3-4, 3-5, 3-6, 3-7, May 1990.
    58. ``Factors Affecting the Growth of Microorganisms in Foods,'' 
The Bad Bug Book, U.S. Food and Drug Administration, Center for Food 
Safety and Applied Nutrition, Foodborne Pathogenic Microorganisms and 
Natural Toxins Handbook, 1997.
    59. FDA ``Guide to Inspections of Computerized Systems in the Food 
Processing Industry,'' Division of Emergency and Investigational 
Operations, Office of Regional Operations, Office of Regulatory 
Affairs, U.S. Food and Drug Administration, March 1998.
    60. Moody, M. W., ``How Cleaning Compounds Do the Job,'' Seafood 
Technology Cooperative Extension Service, Louisiana State University, 
pp. 33-41, undated.
    61. Lentsch, S., ``Sanitizers for an Effective Cleaning Program,'' 
Klenzade Division, Economics Laboratories, St. Paul, MN, undated.
    62. FDA ``Guide to Inspections of Computerized Systems in Drug 
Processing,'' February 1983.
    63. Hitokoto H., S. Morozumi, T. Wauke, S. Sakai, and H. Kurata, 
``Fungal Contamination and Mycotoxin Detection of Powdered Herbal 
Drugs,''

[[Page 12250]]

Applied and Environmental Microbiology, 36:2, pp. 252-256, 1978.
    64. Stewart, M. J., V. Steenkamp, and M. Zuckerman, ``The 
Toxicology of African Herbal Remedies,'' Therapeutic Drug Monitoring, 
20, pp. 510-516, 1998.
    65. Halt, M., ``Moulds and Mycotoxins in Herb Tea and Medicinal 
Plants,'' European Journal of Epidemiology, 14, pp. 269-274, 1998.
    66. Pharmacopeial Previews: Nutritional Supplements--Section 2022: 
Microbial Procedures for Absence of Objectionable Microorganisms in 
Nutritional and Dietary Articles, Pharmacopeial Forum, vol. 25:5, pp. 
8761-8769, September-October 1999.
    67. Pharmacopeial Previews: Nutritional Supplements--Section 2023: 
Microbiological Attributes of Nonsterile Nutritional and Dietary 
Articles, Pharmacopeial Forum, vol. 25:5, pp. 8770-8773, September-
October 1999.
    68. AOAC Official Method 967.22, Vitamin C (Total) in Vitamin 
Preparations, AOAC Official Method 984.26, Vitamin C (Total) in Food, 
Official Methods of Analysis of AOAC International, 16 ed., 
Gaithersburg, MD, 1995.
    69. Madis, V. H. and V. Madis, ``The Sampling of Crude 
Botanicals,'' D&CI, (1986), pp. 38-43 and 79.
    70. Centers for Disease Control, ``Anticholinergic Poisoning 
Associated With an Herbal Tea--New York City, 1994,'' Morbidity and 
Mortality Weekly Report, 44:11, pp. 193-195; March 25, 1995.
    71. Ernst, E., ``Harmless Herbs? A Review of the Recent 
Literature,'' American Journal of Medicine, p. 104, pp. 170-178, 1998.
    72. Huxtable, R. J., ``The Harmful Potential of Herbal and Other 
Plant Products,'' Drug Safety, 5 (Suppl. 1), pp. 126-136, 1990.
    73. Betz, J. M., ``Plant Toxins,'' General Referee Reports: Journal 
of AOAC International, 78:1, pp. 141-144, 1995.
    74. Chan, T. Y. K., J. C. N. Chan, B. Tomlinson, and J. A. J. H. 
Critchley, ``Chinese Herbal Medicines Revisited: A Hong Kong 
Perspective,'' The Lancet, 342, pp. 1532-1534, 1993.
    75. Wong, M. K., P. Tan, and Y. C. Wee, ``Heavy Metals in Some 
Chinese Herbal Plants,'' Biological Trace Element Research, 36, pp. 
135-142, 1993.
    76. ``Reviewer Guidance, Validation of Chromatographic Methods 
Center for Drug Evaluation and Research, FDA, November 1994.
    77. FDA Health Hazard Evaluation, Classification, and FDA 
Enforcement Report for Firm-Initiated Recall #F-321-0, 1989.
    78. FDA Health Hazard Evaluation, Classification, and FDA 
Enforcement Report for Firm-Initiated Recall Vitamin C, U.S. FDA Recall 
#F-657-8, 1998.
    79. FDA Health Hazard Evaluation, Classification, and FDA 
Enforcement Report for Firm-Initiated Recall Vitamin Mineral 
Supplement, U.S. FDA Recall #F-717-0, 1990.
    80. Report of the Commission on Dietary Supplement Labels, Office 
of Disease Prevention and Health Promotion, Department of Health and 
Human Services, Washington, DC 20201, pp. 21 and 45, November 1997.
    81. Webster's II New Riverside University Dictionary, Houghton 
Mifflin Co., 222 Berkeley St., Boston, MA 02116, p. 761, 1994.
    82. ``Manufacturing Practices for Dietary Supplements, 
Pharmacopeial Forum, 28:2, pp. 534-552, March/April 2002.
    83. Draft Standard NSF 173-2001, Dietary Supplements, NSF 
International Draft Standard, NSF International, 789 North Dixboro 
Road, P.O. Box 130140, Ann Arbor, Michigan 48113.
References--Economics
    E1. Research Triangle Institute (RTI), 1999a, ``Economic 
Characterization of the Dietary Supplement Industry,'' Contract No. 
223-96-2290: Task Order 3, March 1999.
    E2. RTI, ``Survey of Manufacturing Practices in the Dietary 
Supplement Industry,'' Contract No. 223-96-2290: Task Order 6, May 
2000.
    E3. NPR/Kaiser Family Foundation/Kennedy School of Government, 
Survey of Americans and Dietary Supplements, 1999.
    E4. Guthrie, J. F., K. M. Koehler, and R. A. Scharff, ``Trends in 
the Consumption of Dietary Supplements Since 1994,'' Unpublished, 2000.
    E5. ``US Dietary Supplements Market Size Expressed as Dollar Sales 
by Top Six Product Categories for 1994 to 1998 and Forecast for 1999 
and 2000,'' Nutrition Business Journal, Dialog File Number 93, San 
Francisco, The Dialog Corp., 2000.
    E6. Supplementary Data Tables, USDA's 1994-96 Continuing Survey of 
Food Intakes by Individuals; Food Surveys Research Group, Beltsville 
Human Nutrition Research Center, Agricultural Research Service, U.S. 
Department of Agriculture, Riverdale, MD; February 1999; accessed at 
http://www.barc.usda.gov/bhnrc/foodsurvey/home.htm. Exit Disclaimer
    E7. Prevention Research Associates, Special Nutritionals Survey, 
Report for Prevention Magazine, June 8, 1999.
    E8. Roe, B. E., B. M. Derby, A. L. Levy, ``Demographic, Lifestyle 
and Information Use Characteristics of Dietary Supplement User 
Segments,'' A report prepared for the Commission on Dietary Supplement 
Labeling, Food and Drug Administration, Center for Food Safety and 
Applied Nutrition, Division of Market Studies, March 12, 1997.
    E9. Eisenberg, D. M., R. B. Davis, S. L. Ettner, S. Appel, S. 
Wildey, M. Van Rompay, R. C. Kessler, ``Trends in Alternative Medicine 
Use in the United States, 1990-1997: Results of a Follow-up National 
Survey,'' Journal of the American Medical Association, 280(18):1569-
1575, 1998.
    E10. Food Marketing Institute, Research Department, and Prevention 
Magazine, Research Department, A Look at the Self Care Movement, Food 
Marketing Institute and Prevention Magazine, 1998.
    E11. Wilde, L. L., ``Information Costs, Duration of Search, and 
Turnover: Theory and Applications,'' Journal of Political Economy, 89, 
December, 1122-1141, 1981.
    E12. McCarthy, P. and B. Timbo, Memorandum: Report of Adverse 
Health Outcomes Associated With the Consumption of Dietary Supplements 
Which Were Prepared Using Poor Manufacturing Practices, November 17, 
1999.
    E13. Prevention Magazine and Food Marketing Institute (FMI), ``The 
Growing Self-Care Movement,'' Prevention/FMI Survey, 2000.
    E14. Gugliotta, G., ``Health Concerns Grow Over Herbal Aids,'' 
Washington Post, March 19, 2000.
    E15. Koehler, K. M., Memorandum: Tally of AERs Regarding Dietary 
Supplements, May 30, 2000.
    E16. Walker, A. M., ``The Relation Between Voluntary Notification 
and Material Risk in Dietary Supplement Safety,'' Harvard School of 
Public Health, March 9, 2000.
    E17. Alsop, R. J., 1999, Wall Street Journal Almanac, New York, 
Ballantine Books, pp. 554-563, 1998.
    E18. Viscusi, W. K., ``The Value of Risks to Life and Health,'' 
Journal of Economic Literature, 31: 1912-1946, 1993.
    E19. Economic Report of the President, Washington, DC, United 
States Government Printing Office, 2000.
    E20. Kaplan, M., J. A. Anderson, and T. G. Ganiats, ``The Quality 
of Well-being Scale: Rationale for a Single Quality of Life Index,'' 
edited by S. R. Walker and R. M. Rosser, In Quality of Life Assessment: 
Key Issues in the 1990's, The Netherlands: Kluwer Academic Publishers, 
65-94, 1993.
    E21. Kaufman, L. D. ``Chronicity of the Eosinophilia-Myalgia 
Syndrome. A

[[Page 12251]]

Reassessment After Three Years,'' Arthritis and Rheumatism, 37:884-887, 
1994 Abstract.
    E22. Hedberg, C., et al., ``Eosinophilia-Myalgia Syndrome, Natural 
History in a Population Based Cohort,'' Archives of Internal Medicine, 
152:1189-1892, 1992 Abstract.
    E23. Vanherweghem, J. L., ``Misuse of Herbal Remedies: The Case of 
an Outbreak of Terminal Renal Failure in Belgium (Chinese Herbs 
Nephropathy),'' Journal of Alternative Complementary Medicine, 4:9-13, 
1998.
    E24. Jarrell, G. and S. Peltzman, ``Impact of Product Recalls on 
the Wealth of Sellers,'' Journal of Political Economy, 93:512-536, 
1985.
    E25. ``Herbal Rx: The Promises and Pitfalls,'' Consumer Reports, 
March, 44-48, 1999.
    E26. ConsumerLab, Independent tests of herbal, vitamin, and mineral 
supplements, accessed at http://www.consumerlab.com/results Exit Disclaimer on March 
16, 2000.
    E27. Rothman, G., ``Herbal Remedy Ripoffs,'' D Magazine, April, 39-
44, 2000.
    E28. Foster, W. and R. E. Just, ``Measuring Welfare Effects of 
Product Contamination with Consumer Uncertainty,'' Journal of 
Environmental Economics and Management, 17, pp. 266-283, 1989.
    E29. Jones, P. and J. Hudson, ``Standardization and the Costs of 
Assessing Quality,'' European Journal of Political Economy, 12:355-361, 
1996.
    E30. Newman, J. W. and B. D. Lockeman, ``Measuring Prepurchase 
Information Seeking,'' Journal of Consumer Research, 2:216-222, 1975.
    E31. Punj, G. N. and R. Staelin, ``A Model of Consumer Information 
Search Behavior for New Automobiles,'' Journal of Consumer Research, 
9:366-380, 1983.
    E32. Duncan C. and R. Olshavsky, ``External Search: The Role of 
Consumer Beliefs,'' Journal of Marketing Research, 19:32-43, 1982.
    E33. Andrews, R. L., ``Economics of Information and Heterogeneous 
Products,'' Journal of Economic Psychology, 13:399-420, 1992.
    E34. Avery, R. J., ``Determinants of Search for Nondurable Goods: 
An Empirical Assessment of the Economics of Consumer Theory,'' The 
Journal of Consumer Affairs, 30(2):390-420, 1996.
    E35. Hawkins, D. I., R. J. Best, and K. A. Coney, Consumer 
Behavior: Implications for Marketing Strategy, 6th ed., Chicago: Irwin, 
1995.
    E36. ``The Power of Persuasion at the Moment of Truth,'' Drug Store 
News, 19(20):3-8, 22, 24, 1997.
    E37. Robinson, J. P. and G. Godbey, Time for Life: The Surprising 
Ways Americans Use Their Time, University Park, PA: The Pennsylvania 
State University Press, 1997.
    E38. ``How Consumers Shop,'' Progressive Grocer, December, 62-64, 
1992.
    E39. United States Department of Labor, Bureau of Labor Statistics 
(BLS), http://stats.bls.gov/csx/1998/Standard/cucomp.pdf, Exit Disclaimer as obtained 
on December 8, 1999.
    E40. Lippmann S. and J. McCall, ``The Economics of Job Search,'' 
Journal of Economic Theory, 14:155-189, 1976.
    E41. Little, W. R., ``Herbal Products: A Retail Pharmacists 
Perspective,'' Nutraceuticals World 3 (May/June): 58-66, 2000.
    E42. Bureau of Labor Statistics (BLS), ``Consumer Price Index-All 
Urban Consumers,'' Accessed at http://146.142.4.24/cgi-bin/
surveymost?cu Exit Disclaimer
 on July 13, 1999.
    E43. Gurley, B. J. , S. F. Gardner, and M. A. Hubbard, ``Content 
Versus Label Claims in Ephedra-Containing Dietary Supplements,'' 
American Journal Health Systems Pharmacy, 57:963-969, May 15, 2000.
    E44. Eastern Research Group (ERG), ``Economic Analysis of Proposed 
Revisions to the Good Manufacturing Practices Regulation for Medical 
Devices,'' Contract No. 223-91-8100: Task Order 1, pp. 4-23 to 4-25, 
November, 1993.
    E45. R. S. Means, Building Construction Cost Data, 2000.
    E46. RTI, 1999b, ``Dietary Supplement Sales Information,'' Contract 
No. 223-96-2290: Task Order 4, October, 1999.
    E47. Vardon, P., Memorandum: Conversation with industry expert 
regarding inventory practices, April 4, 2000.
    E48. Ramey, V. A. and K. West, ``Inventories,'' National Bureau of 
Economic Research, working paper 6315, Abstract, 1997.
    E49. Madis, V. H. and V. Madis, ``The Sampling of Crude 
Botanicals,'' Drug and Cosmetic Industry, pp. 38-79, 1986.
    E50. Plant Bioactives Research Institute, ``Costs for Analysis,'' 
accessed at http://www.plant-bioactives.com Exit Disclaimer on April 10, 2000.
    E51. ``Fees, Methods, and Analytical Capabilities of Laboratories-
Summary,'' unpublished summary of prices listed on the Internet, 1999.
    52. McLaughlin, C., Memorandum: Costs of pest control, 2000.

List of Subjects

21 CFR Part 111

    Dietary foods, Drugs, Foods, Packaging and containers.

21 CFR Part 112

    Drugs, Packaging and containers, Labeling.
    Therefore, under the Federal Food, Drug, and Cosmetic Act and under 
authority delegated to the Commissioner of Food and Drugs, FDA proposes 
to amend 21 CFR chapter I, parts 111 and 112 as set forth below:

PART 111--CURRENT GOOD MANUFACTURING PRACTICE IN MANUFACTURING, 
PACKING, OR HOLDING DIETARY INGREDIENTS AND DIETARY SUPPLEMENTS

    1. The authority citation for part 111 is revised to read as 
follows:

    Authority: 21 U.S.C. 321, 342, 343, 348, 371, 374, 381, 393; 42 
U.S.C. 264.

    2. The part heading for part 111 is revised as set forth above.
    3. Add new subpart A to part 111 to read as follows:
Subpart A--General Provisions
Sec.
111.1 Who is subject to these regulations?
111.2 What are these regulations intended to accomplish?
111.3 What definitions apply to this part?
111.5 Do other statutory provisions and regulations apply?
111.6 Exclusions.

Subpart A--General Provisions

Sec.  111.1  Who is subject to these regulations?

    You are subject to the regulations in this part if you manufacture, 
package, or hold a dietary ingredient or dietary supplement.

Sec.  111.2  What are these regulations intended to accomplish?

    The regulations in this part establish the minimum current good 
manufacturing practices that you must use to the extent that you 
manufacture, package, or hold a dietary ingredient or dietary 
supplement.

Sec.  111.3  What definitions apply to this part?

    The definitions and interpretations of terms in section 201 of the 
Federal Food, Drug, and Cosmetic Act (the act) apply to such terms when 
used in these regulations. For the purpose of these regulations, the 
following definitions also apply:
    Actual yield means the quantity that is actually produced at any 
appropriate step of manufacture or packaging of a particular dietary 
ingredient or dietary supplement.
    Batch means a specific quantity of a dietary ingredient or dietary 
supplement that is intended to meet specifications for identity, 
purity, quality, strength,

[[Page 12252]]

and composition, and is produced during a specified time period 
according to a single manufacturing record during the same cycle of 
manufacture.
    Batch number, lot number, or control number means any distinctive 
group of letters, numbers, or symbols, or any combination of them, from 
which the complete history of the manufacturing, packaging, or holding 
of a batch or lot of dietary ingredients or dietary supplements can be 
determined.
    Component means any substance intended for use in the manufacture 
of a dietary ingredient or dietary supplement including those that may 
not appear in the finished dietary ingredient or dietary supplement. 
Component includes ingredients and dietary ingredients as described in 
section 201(ff) of the Act.
    Consumer complaint means communication that contains any 
allegation, written or oral, expressing dissatisfaction with the 
quality of a dietary ingredient or a dietary supplement related to good 
manufacturing practices. Examples of product quality related to good 
manufacturing practices are: Foul odor, off taste, superpotent, 
subpotent, wrong ingredient, drug contaminant, other contaminant (e.g., 
bacteria, pesticide, mycotoxin, glass, lead), disintegration time, 
color variation, tablet size or size variation, under-filled container, 
foreign material in a dietary supplement container, improper packaging, 
or mislabeling. For the purposes of the regulations in this part, a 
consumer complaint about product quality may or may not include 
concerns about a possible hazard to health. However, a consumer 
complaint does not include an adverse event, illness, or injury related 
to the safety of a particular dietary ingredient independent of whether 
the product is produced under good manufacturing practices.
    Contact surface means any surface that contacts a component, 
dietary ingredient, dietary supplement, and those surfaces from which 
drainage onto the component, dietary ingredient, dietary supplement, or 
onto surfaces that contact the component, dietary ingredient, or 
dietary supplement ordinarily occurs during the normal course of 
operations. Examples of contact surfaces include, but are not limited 
to, containers, utensils, tables, contact surfaces of equipment, and 
packaging.
    Ingredient means any substance that is used in the manufacture of a 
dietary ingredient or dietary supplement that is intended to be present 
in the finished dietary ingredient or dietary supplement. An ingredient 
includes, but is not necessarily limited to, a dietary ingredient as 
described in section 201(ff) of the Act.
    Inprocess material means any material that is fabricated, 
compounded, blended, ground, extracted, sifted, sterilized, derived by 
chemical reaction, or processed in any other way for use in the 
manufacture of a dietary ingredient or dietary supplement.
    Lot means a batch, or a specific identified portion of a batch 
intended to have uniform identity, purity, quality, strength, and 
composition; or, in the case of a dietary ingredient or dietary 
supplement produced by continuous process, a specific identified amount 
produced in a specified unit of time or quantity in a manner that is 
intended to have uniform identity, purity, quality, strength, and 
composition.
    Microorganisms means yeasts, molds, bacteria, viruses, and other 
similar microscopic organisms having public health or sanitary concern. 
This definition includes, but is not limited to, species that:
    (1) Have public health significance;
    (2) Could cause a component, dietary ingredient, or dietary 
supplement to decompose;
    (3) Indicate that the component, dietary ingredient, or dietary 
supplement is contaminated with filth; or
    (4) Otherwise may cause the component, dietary ingredient, or 
dietary supplement to be adulterated.
    Must is used to state mandatory requirements.
    Pest means any objectionable insects or other animals including, 
but not limited to, birds, rodents, flies, mites, and larvae.
    Physical plant means all or parts of a building or facility used 
for or in connection with manufacturing, packaging, or holding a 
dietary ingredient or dietary supplement.
    Quality control means a planned and systematic operation or 
procedure for preventing a dietary ingredient or dietary supplement 
from being adulterated.
    Quality control unit means any person or group that you designate 
to be responsible for quality control operations.
    Representative sample means a sample that consists of a number of 
units that are drawn based on rational criteria, such as random 
sampling, and intended to ensure that the sample accurately portrays 
the material being sampled.
    Reprocessing means using, in the manufacture of a dietary 
ingredient or a dietary supplement, clean, unadulterated components, 
dietary ingredients, or dietary supplements that have been previously 
removed from manufacturing for reasons other than insanitary conditions 
and that have been made suitable for use in the manufacture of a 
dietary ingredient or dietary supplement.
    Sanitize means to adequately treat equipment, containers, utensils, 
or any other dietary product contact surface by applying cumulative 
heat or chemicals on cleaned food contact surfaces that when evaluated 
for efficacy, yield a reduction of 5 logs, which is equal to 99.999 
percent reduction, of representative disease microorganisms of public 
health significance and substantially reduce the numbers of other 
undesirable microorganisms, but without adversely affecting the product 
or its safety for the consumer.
    Theoretical yield means the quantity that would be produced at any 
appropriate step of manufacture or packaging of a particular dietary 
ingredient or dietary supplement, based upon the quantity of components 
or packaging to be used, in the absence of any loss or error in actual 
production.
    Water activity (aw) is a measure of the free moisture in 
a component, dietary ingredient, or dietary supplement and is the 
quotient of the water vapor pressure of the substance divided by the 
vapor pressure of pure water at the same temperature.
    We means the United States Food and Drug Administration (FDA).
    You means a person who manufactures, packages, or holds dietary 
ingredients or dietary supplements.

Sec.  111.5  Do other statutory provisions and regulations apply?

    In addition to the regulations in this part, you must comply with 
other applicable statutory provisions and regulations under the Act 
related to the manufacturing, packaging, or holding of dietary 
ingredients or dietary supplements.

Sec.  111.6  Exclusions.

    The regulations in this part do not apply to a person engaged 
solely in activities related to the harvesting, storage, or 
distribution of raw agricultural commodities that will be incorporated 
into a dietary ingredient or dietary supplement by other persons.
    4. Add new subpart B to part 111 to read as follows:
Subpart B--Personnel
Sec.
111.10 What microbial contamination and hygiene requirements apply?
111.12 What personnel qualification requirements apply?

[[Page 12253]]

111.13 What supervisor requirements apply?

Subpart B--Personnel

Sec.  111.10  What microbial contamination and hygiene requirements 
apply?

    (a) Microbial contamination. You must take measures to exclude from 
any operations any person who might be a source of microbial 
contamination of any material including components, dietary 
ingredients, dietary supplements, and contact surfaces used in the 
manufacture, packaging, or holding of a dietary ingredient or a dietary 
supplement. Such measures include, but are not limited to, the 
following:
    (1) Excluding any person who, by medical examination or supervisory 
observation, is shown to have, or appears to have an illness, open 
lesion, or any other abnormal source of microbial contamination, which 
may be expected to result in microbial contamination of components, 
dietary ingredients, dietary supplements, or contact surfaces, from 
working in any operations until the condition is corrected; and
    (2) Instructing your employees to notify their supervisor(s) if 
they have or if there is a reasonable possibility that they have a 
health condition described in paragraph (a)(1) of this section that 
could contaminate any components, dietary ingredients, dietary 
supplements, or any contact surface.
    (b) Hygienic practices. If you work in operations during which 
adulteration of the component, dietary ingredients, dietary supplement, 
or contact surface may occur, you must use hygienic practices to the 
extent necessary to protect against contamination of components, 
dietary ingredients, dietary supplements, or contact surfaces. These 
hygienic practices include, but are not limited to:
    (1) Wearing outer garments in a manner that protects against the 
contamination of components, dietary ingredients, dietary supplements, 
or any contact surface;
    (2) Maintaining adequate personal cleanliness;
    (3) Washing hands thoroughly (and sanitizing if necessary to 
protect against contamination with microorganisms) in an adequate hand-
washing facility:
    (i) Before starting work; and
    (ii) At any time when the hands may have become soiled or 
contaminated;
    (4) Removing all unsecured jewelry and other objects that might 
fall into components, dietary ingredients, dietary supplements, 
equipment, or packaging, and removing hand jewelry that cannot be 
adequately sanitized during periods in which components, dietary 
ingredients, or dietary supplements are manipulated by hand. If hand 
jewelry cannot be removed, it must be covered by material that is 
maintained in an intact, clean, and sanitary condition and that 
effectively protects against contamination of components, dietary 
ingredients, dietary supplements, or contact surfaces;
    (5) Maintaining gloves used in handling components, dietary 
ingredients, or dietary supplements in an intact, clean, and sanitary 
condition. The gloves must be of an impermeable material;
    (6) Wearing, where appropriate, in an effective manner, hair nets, 
caps, beard covers, or other effective hair restraints;
    (7) Not storing clothing or other personal belongings in areas 
where components, dietary ingredients, or dietary supplements or any 
contact surfaces are exposed or where contact surfaces are washed;
    (8) Not eating food, chewing gum, drinking beverages and using 
tobacco products in areas where components, dietary ingredients, 
dietary supplements, or any contact surfaces are exposed, or where 
contact surfaces are washed; and
    (9) Taking any other precautions necessary to protect against the 
contamination of components, dietary ingredients, dietary supplements, 
or contact surfaces with microorganisms, filth, or any other extraneous 
materials, including, but not limited to, perspiration, hair, 
cosmetics, tobacco, chemicals, and medicines applied to the skin.

Sec.  111.12  What personnel qualification requirements apply?

    (a) You must have qualified employees to manufacture, package, or 
hold dietary ingredients or dietary supplements; and
    (b) Each person engaged in manufacturing, packaging, or holding 
must have the training and experience to perform the person's duties.

Sec.  111.13  What supervisor requirements apply?

    (a) You must assign qualified personnel to supervise the 
manufacturing, packaging, or holding of dietary ingredients and dietary 
supplements.
    (b) You and the supervisors you use must be qualified by training 
and experience to supervise.
    5. Add new subpart C to part 111 to read as follows:
Subpart C--Physical Plant
Sec.
111.15 What sanitation requirements apply to your physical plant?
111.20 What design and construction requirements apply to your 
physical plant?

Subpart C--Physical Plant

Sec.  111.15  What sanitation requirements apply to your physical 
plant?

    (a) Physical plant facilities. (1) You must maintain your physical 
plant in a clean and sanitary condition; and
    (2) You must keep your physical plant in repair sufficient to 
prevent components, dietary ingredients, dietary supplements, or 
contact surfaces from becoming contaminated.
    (b) Cleaning compounds, sanitizing agents, and pesticides. (1) You 
must use cleaning compounds and sanitizing agents that are free from 
microorganisms of public health significance and safe and adequate 
under the conditions of use.
    (2) You must not use or hold toxic materials in a physical plant in 
which contact surfaces, components, dietary ingredients, or dietary 
supplements are manufactured or exposed, unless those materials are 
necessary:
    (i) To maintain clean and sanitary conditions;
    (ii) For use in laboratory testing procedures;
    (iii) For maintaining or operating the physical plant or equipment; 
or
    (iv) For use in the plant's operations.
    (3) You must identify and hold toxic cleaning compounds, sanitizing 
agents, pesticides, and pesticide chemicals in a manner that protects 
against contamination of components, dietary ingredients, dietary 
supplements, or contact surfaces.
    (c) Pest control. (1) You must not allow animals or pests in any 
area of your physical plant. Guard or guide dogs are allowed in some 
areas of your physical plant if the presence of the dogs will not 
result in contamination of components, dietary ingredients, dietary 
supplements, or contact surfaces;
    (2) You must take effective measures to exclude pests from the 
physical plant and to protect against contamination of components, 
dietary ingredients, dietary supplements, and contact surfaces on the 
premises by pests; and
    (3) You must not use insecticides, fumigants, fungicides, or 
rodenticides, unless you take precautions to protect against the 
contamination of components, dietary ingredients, dietary supplements, 
or contact surfaces.
    (d) Water supply. (1) You must provide water that is safe and of 
adequate sanitary quality, at suitable temperatures, and under pressure 
as

[[Page 12254]]

needed, in all areas where water is necessary for:
    (i) Manufacturing dietary ingredients or dietary supplements;
    (ii) Making ice that comes in contact with components, dietary 
ingredients, dietary supplements, or contact surfaces;
    (iii) Cleaning any surface; and
    (iv) Employee bathrooms and hand-washing facilities.
    (2) Water that contacts components, dietary ingredients, dietary 
supplements, or any contact surface must at a minimum comply with the 
National Primary Drinking Water regulations prescribed by the 
Environmental Protection Agency under 40 CFR part 141 and any state and 
local government requirements;
    (3) You must have documentation or otherwise be able to show that 
water that contacts components, dietary ingredients, dietary 
supplements, or any contact surface meets the requirements in paragraph 
(d)(2) of this section.
    (e) Plumbing. The plumbing in your physical plant must be of an 
adequate size and design and be adequately installed and maintained to:
    (1) Carry sufficient amounts of water to required locations 
throughout the physical plant;
    (2) Properly convey sewage and liquid disposable waste from your 
physical plant;
    (3) Avoid being a source of contamination to components, dietary 
ingredients, dietary supplements, water supplies, or any contact 
surface, or creating an unsanitary condition;
    (4) Provide adequate floor drainage in all areas where floors are 
subject to flooding-type cleaning or where normal operations release or 
discharge water or other liquid waste on the floor; and
    (5) Not allow backflow from, or cross connection between, piping 
systems that discharge waste water or sewage and piping systems that 
carry water used for manufacturing dietary ingredients or dietary 
supplements, for cleaning contact surfaces, or for use in bathrooms or 
hand-washing facilities.
    (f) Sewage disposal. You must dispose of sewage into an adequate 
sewage system or through other adequate means.
    (g) Bathrooms. You must provide your employees with adequate, 
readily accessible bathrooms. The bathrooms must be kept clean and must 
not become a potential source of contamination to components, dietary 
ingredients, dietary supplements, or contact surfaces. You must:
    (1) Keep the bathrooms in good repair at all times;
    (2) Provide self-closing doors; and
    (3) Provide doors that do not open into areas where components, 
dietary ingredients, dietary supplements, or contact surfaces are 
exposed to airborne contamination except where alternate means have 
been taken to protect against contamination (such as double doors or 
positive airflow systems).
    (h) Hand-washing facilities. You must provide hand-washing 
facilities that are adequate, convenient, and furnish running water at 
a suitable temperature. You must do this by providing:
    (1) Hand-washing and, where appropriate, hand-sanitizing facilities 
at each location in your physical plant where good hygienic practices 
require employees to wash or to sanitize or both wash and sanitize 
their hands;
    (2) Effective hand-cleaning and sanitizing preparations;
    (3) Air driers, sanitary towel service, such as disposable paper 
towels, or other suitable drying devices;
    (4) Devices or fixtures, such as water control valves, designed and 
constructed to protect against recontamination of clean, sanitized 
hands;
    (5) Signs that are easy to understand and are posted throughout the 
physical plant that direct employees handling components, dietary 
ingredients, dietary supplements, or contact surfaces to wash and, 
where appropriate, to sanitize their hands before they start work, 
after each absence from their duty station, and when their hands may 
have become soiled or contaminated; and
    (6) Trash bins that are constructed and maintained in a manner to 
protect against recontamination of hands and contamination of 
components, dietary ingredients, dietary supplements, or any contact 
surface.
    (i) Trash disposal. You must convey, store, and dispose of trash 
to:
    (1) Minimize the development of odor;
    (2) Minimize the potential for the trash to attract, harbor, or 
become a breeding place for pests;
    (3) Protect against contamination of components, dietary 
ingredients, dietary supplements, any contact surface, water supplies, 
and grounds surrounding your physical plant; and
    (4) Control hazardous waste to prevent contamination of components, 
dietary ingredients, dietary supplements, and contact surfaces.
    (j) Sanitation supervisors. You must assign one or more employees 
to supervise overall sanitation. These supervisors must be qualified by 
training and experience to develop and supervise sanitation procedures.

Sec.  111.20  What design and construction requirements apply to your 
physical plant?

    Any physical plant you use in the manufacture, packaging, or 
holding of dietary ingredients or dietary supplements must:
    (a) Be suitable in size, construction, and design to facilitate 
maintenance, cleaning, and sanitizing operations;
    (b) Have adequate space for the orderly placement of equipment and 
holding materials as is necessary for maintenance, cleaning, and 
sanitizing operations and to prevent contamination and mixups of 
components, dietary ingredients, and dietary supplements during 
manufacturing, packaging, or holding;
    (c) Permit the use of proper precautions to reduce the potential 
for mixups or contamination of components, dietary ingredients, dietary 
supplements, or contact surfaces, with microorganisms, chemicals, 
filth, or other extraneous material. Your physical plant must have and 
you must use separate or defined areas of adequate size or other 
control systems, such as computerized inventory controls or automated 
systems of separation, to prevent contamination and mixups of 
components, dietary ingredients, and dietary supplements during the 
following operations:
    (1) Receiving, identifying, holding, and withholding from use, 
components, dietary ingredients, dietary supplements, packaging, and 
labels that will be used in or during the manufacturing, packaging, or 
holding of dietary ingredients and dietary supplements;
    (2) Separating, as necessary, components, dietary ingredients, 
dietary supplements, packaging, and labels that are to be used from 
components, dietary ingredients, dietary supplements, packaging, or 
labels that are awaiting material review and disposition decision, 
reprocessing, or are awaiting disposal after rejection;
    (3) Separating the manufacturing, packaging, and holding of 
different product types including, but not limited to, different types 
of dietary ingredients, dietary supplements and other foods, cosmetics, 
and pharmaceutical products;
    (4) Performing laboratory analyses and holding laboratory supplies 
and samples;
    (5) Cleaning and sanitizing contact surfaces;
    (6) Packaging and label operations; and
    (7) Holding dietary ingredients or dietary supplements.
    (d) Be designed and constructed in a manner that prevents 
contamination of

[[Page 12255]]

components, dietary ingredients, dietary supplements, or contact 
surfaces. The design and construction must include, but not be limited 
to:
    (1) Floors, walls, and ceilings that are of smooth and hard 
surfaces that can be adequately cleaned and kept clean and in good 
repair;
    (2) Fixtures, ducts, and pipes that do not contaminate components, 
dietary ingredients, dietary supplements, or contact surfaces by 
dripping or condensate;
    (3) Adequate ventilation or environmental control equipment such as 
air flow systems, including filters, fans, and other air-blowing 
equipment, that minimize odors and vapors (including steam and noxious 
fumes) in areas where they may contaminate components, dietary 
ingredients, dietary supplements, or contact surfaces;
    (4) Fans and other air-blowing equipment located and operated in a 
manner that minimizes the potential for microorganisms and particulate 
matter to contaminate components, dietary ingredients, dietary 
supplements, or contact surfaces;
    (5) Equipment that controls temperature and humidity; and
    (6) Aisles or working spaces between equipment and walls that are 
adequately unobstructed and of adequate width to permit all persons to 
perform their duties and to protect against contamination of 
components, dietary ingredients, dietary supplements, or contact 
surfaces with clothing or personal contact.
    (e) Provide adequate light in:
    (1) All areas where components, dietary ingredients, or dietary 
supplements are examined, processed, or held;
    (2) All areas where contact surfaces are cleaned; and
    (3) Hand-washing areas, dressing and locker rooms, and bathrooms.
    (f) Use safety-type light bulbs, fixtures, skylights, or other 
glass that is suspended over exposed components, dietary ingredients, 
or dietary supplements in any step of preparation, unless otherwise 
constructed in a manner that will protect against contamination of 
components, dietary ingredients, or dietary supplements in case of 
glass breakage.
    (g) Provide protection by any effective means against contamination 
of components, dietary ingredients, and dietary supplements in bulk 
fermentation vessels, including consideration of:
    (1) Use of protective coverings;
    (2) Placement in areas where you can eliminate harborages for pests 
over and around the vessels;
    (3) Placement in areas where you can check regularly for pests, 
pest infestation, filth or any other extraneous materials; and
    (4) Use of skimming equipment.
    (h) Use adequate screening or other protection against pests, where 
necessary.
    6. Add new subpart D to part 111 to read as follows:
Subpart D--Equipment and Utensils
Sec.
111.25 What requirements apply to the equipment and utensils you 
use?
111.30 What requirements apply to automatic, mechanical, or 
electronic equipment?

Subpart D--Equipment and Utensils

Sec.  111.25  What requirements apply to the equipment and utensils you 
use?

    (a)(1) You must use equipment and utensils that are of appropriate 
design, construction, and workmanship to enable them to be suitable for 
their intended use and to be adequately cleaned and properly 
maintained. Equipment and utensils include, but are not limited to, the 
following:
    (i) Equipment used to hold or convey;
    (ii) Equipment used to measure;
    (iii) Equipment using compressed air or gas;
    (iv) Equipment used to carry out processes in closed pipes and 
vessels; and
    (v) Equipment used in automatic, mechanical, or electronic systems.
    (2) You must use equipment and utensils of appropriate design and 
construction so that use will not result in the contamination of 
components, dietary ingredients, or dietary supplements with:
    (i) Lubricants;
    (ii) Fuel;
    (iii) Coolants;
    (iv) Metal or glass fragments;
    (v) Filth or any other extraneous material;
    (vi) Contaminated water; or
    (vii) Any other contaminants.
    (3) All equipment and utensils you use must be:
    (i) Installed and maintained to facilitate cleaning the equipment, 
utensils, and all adjacent spaces;
    (ii) Corrosion-resistant if the equipment or utensils contact 
components, dietary ingredients, or dietary supplements;
    (iii) Made of nontoxic materials;
    (iv) Designed and constructed to withstand the environment of their 
intended use, the action of components, dietary ingredients, or dietary 
supplements, and, if applicable, cleaning compounds and sanitizing 
agents; and
    (v) Maintained to protect components, dietary ingredients, and 
dietary supplements from being contaminated by any source.
    (4) Equipment and utensils you use must have seams that are 
smoothly bonded or maintained to minimize accumulation of component, 
dietary ingredient, or dietary supplement particles, dirt, filth, 
organic material, or any other extraneous materials or contaminants to 
minimize the opportunity for growth of microorganisms.
    (5) Each freezer and cold storage compartment you use to hold 
components, dietary ingredients, or dietary supplements:
    (i) Must be fitted with an indicating thermometer, temperature-
measuring device, or temperature-recording device that shows the 
temperature accurately within the compartment; and
    (ii) Must have an automatic device for regulating temperature or an 
automatic alarm system to indicate a significant temperature change in 
a manual operation.
    (6) Instruments or controls used in the manufacturing, packaging, 
or holding of a dietary ingredient or dietary supplement, including but 
not limited to, instruments or controls you use to measure, regulate, 
or record temperatures, hydrogen ion concentration (pH), water 
activity, or other conditions that control or prevent the growth of 
microorganisms or other contamination must be:
    (i) Accurate and precise;
    (ii) Adequately maintained; and
    (iii) Adequate in number for their designated uses.
    (7) Compressed air or other gases you introduce mechanically into 
or onto a component, dietary ingredient, dietary supplement, or contact 
surface or that you use to clean any contact surface must be treated in 
such a way that the component, dietary ingredient, dietary supplement, 
or contact surface is not contaminated.
    (b)(1) You must calibrate instruments and controls you use in 
manufacturing or testing a component, dietary ingredient, or dietary 
supplement.
    (2) You must calibrate before first use; and
    (i) As specified in writing by the manufacturer of the instrument 
and control, or
    (ii) At routine intervals or as otherwise necessary to ensure the 
accuracy and precision of the instrument and control.
    (c) You must:
    (1) Establish a written procedure for calibrating instruments and 
controls you use in manufacturing or testing a

[[Page 12256]]

component, dietary ingredient, or dietary supplement and document that 
the written procedure was followed each time a calibration is 
performed, or
    (2) Document at the time of performance that the instrument and 
control calibration established in accordance with this section was 
performed.
    (d) You must identify the following for calibrating instruments and 
controls in any written procedure or at the time of performance:
    (1) The instrument or control calibrated;
    (2) The date of calibration;
    (3) The reference standard used including the certification of 
accuracy of the known reference standard and a history of 
recertification of accuracy;
    (4) The calibration method used including appropriate limits for 
accuracy and precision of instruments and controls when calibrating;
    (5) The calibration reading or readings found; and
    (6) The recalibration method used if accuracy or precision or both 
accuracy and precision limits for instruments and controls were not 
met; and
    (7) The initials of the person who performed the calibration.
    (d) You must repair or replace instruments or controls that cannot 
be adjusted to agree with the reference standard.
    (e)(1) You must maintain, clean, and sanitize as necessary, all 
equipment, utensils, and any other contact surfaces that are used to 
manufacture, package, or hold components, dietary ingredients, or 
dietary supplements. Equipment and utensils must be taken apart as 
necessary for thorough maintenance, cleaning, and sanitizing.
    (2) You must ensure that all contact surfaces used for 
manufacturing or holding of low-moisture components, dietary 
ingredients, or dietary supplements are in a dry and sanitary condition 
at the time of use. When the surfaces are wet-cleaned, they must be 
sanitized, when necessary, and thoroughly dried before subsequent use.
    (3) If you use wet processing during manufacturing, you must clean 
and sanitize all contact surfaces, as necessary, to protect against the 
introduction of microorganisms into components, dietary ingredients, or 
dietary supplements. When cleaning and sanitizing is necessary, you 
must clean and sanitize all contact surfaces before use and after any 
interruption during which the contact surface may have become 
contaminated. If you use contact surfaces in a continuous production 
operation or in back-to-back operations involving different batches of 
the same dietary ingredient or dietary supplement, you must clean and 
sanitize the contact surfaces as necessary.
    (4) You must clean surfaces that do not touch components, dietary 
ingredients, or dietary supplements as frequently as necessary to 
protect against contaminating components, dietary ingredients, or 
dietary supplements.
    (5) Single-service articles (such as utensils intended for one-time 
use, paper cups, and paper towels) must be:
    (i) Stored in appropriate containers; and
    (ii) Handled, dispensed, used, and disposed of in a manner that 
protects against contamination of components, dietary ingredients, 
dietary supplements, or any contact surface.
    (6) Cleaning compounds and sanitizing agents must be adequate for 
intended use and safe under condition of use;
    (7) You must store cleaned and sanitized portable equipment and 
utensils that have contact surfaces in a location and manner that 
protects them from contamination.
    (f) You must keep calibration records as required by this section 
in accordance with Sec.  111.125.

Sec.  111.30  What requirements apply to automatic, mechanical, or 
electronic equipment?

    (a) When you use automatic, mechanical, or electronic equipment to 
manufacture, package, label, and hold a dietary ingredient or dietary 
supplement, you must:
    (1) Design or select equipment to ensure that dietary ingredient or 
dietary supplement specifications are consistently achieved and
    (2) Determine the suitability of your equipment by ensuring that 
your equipment is capable of operating satisfactorily within the 
operating limits required by the process.
    (b) For any automatic, mechanical, or electronic equipment you use, 
you must:
    (1) Routinely calibrate, inspect, or check to ensure proper 
performance. Your quality control unit must approve these calibrations, 
inspections, or checks;
    (2) Make and keep written records of equipment calibrations, 
inspections, or checks;
    (3) Establish and use appropriate controls, to ensure that your 
quality control unit approves changes in the master manufacturing 
record, batch control records, packaging operations and label 
operations, or changes to other operations related to the equipment 
that you use and that only authorized personnel institute the changes;
    (4) Establish and use appropriate controls to ensure that the 
equipment functions in accordance with its intended use. These controls 
must be approved by your quality control unit; and
    (5) Make and keep backup file(s) of software programs and of data 
entered into your computer system. Your backup file (e.g., a hard copy 
of data you have entered, diskettes, tapes, microfilm, or compact 
disks) must be an exact and complete record of the data you entered. 
You must keep your backup software programs and data secure from 
alterations, inadvertent erasures, or loss.
    (c) You must keep automatic, mechanical, or electronic equipment 
records required by this section in accordance with Sec.  111.125.

Sec.  111.50  [Redesignated as Sec.  111.72 and Amended]

    7. Redesignate Sec.  111.50 as Sec.  111.72 and transfer it to a 
new subpart E, Production and Process Controls, and revise the section 
heading to read as follows:

Sec.  111.72  What requirements apply to packaging of iron-containing 
dietary supplements?

* * * * *
    8. Add Sec. Sec.  111.35 through 111.70 and Sec.  111.74 to newly 
added subpart E to read as follows:

Sec.  111.35  What production and process controls must you use?

    (a) You must implement a system of production and process controls 
that covers all stages of manufacturing, packaging, labeling, and 
holding of the dietary ingredients and dietary supplements.
    (b) Your production and in-process control system must be designed 
to ensure that the dietary ingredient or dietary supplement is 
manufactured, packaged, and held in a manner that will prevent 
adulteration of the dietary ingredient or dietary supplement. The 
production and in-process control system must include all requirements 
of this subpart and must be reviewed and approved by the quality 
control unit.
    (c) You must use a quality control unit in your manufacturing, 
packaging, and label operations for producing the dietary ingredient or 
dietary supplement to ensure that these operations are performed in a 
manner that prevents adulteration and ensures that the dietary 
ingredient or dietary supplement meets specifications for identity, 
purity, quality, strength, and composition.

[[Continued on page 12257]]

 
 

Local Navigation

Jump to main content.