Brief Filed with the FDA on Behalf of Supplement Manufacturers The Battle for Life he FDA’s Current Good Manufacturing Practice (CGMP) proposal for dietary supplements is not really about quality control—it’s about control of the marketplace. CGMP hands the FDA the power to do everything it ever wanted, under the rationale that poor standards are rampant. Yet if CGMP is accepted in any form even remotely similar to that of the current proposal, the industry will be destroyed. Only the largest firms will remain—several of which are already controlled by the pharmaceutical industry—and innovation will fall off dramatically. Most complex formulations will disappear, many supplements will become unavailable, permissible serving sizes will decrease, and prices will skyrocket. To help prevent the death of our industry, Life Enhancement Products contributed to the preparation of the Rubin Economic Impact Assessment of CGMP, referred to in the brief. Dr. Rubin found that the annual benefits of CGMP are overstated by more than 15 times, and the costs of compliance are understated by 10 times. These are gross errors! The costs of compliance for the small companies that constitute about 80% of the industry would be so great that few would survive. CGMP reminds us of that infamous quote from the Tet Offensive in the Vietnam War: “We had to destroy the town in order to save it.” Insanity! For the complete report, go to http://emord.com/; it is an exhibit to the CGMP brief filed by Emord and Associates, the attorneys for Durk Pearson & Sandy Shaw. Before the DEPARTMENT OF HEALTH AND HUMAN SERVICES Food and Drug Administration Rockville, MD In re: Current Good Manufacturing Practice in Manufacturing Packaging, or Holding Dietary Ingredients and Dietary Supplements ) ) ) ) ) Docket No. 96N-0417 August 11, 2003 COMMENTS OF DURK PEARSON & SANDY SHAW Durk Pearson & Sandy Shaw (Commenters) hereby submit their comments in response to the proposed rule in the above-referenced docket, 68 Federal Register 12158 (13 March 2003) (hereinafter the “Proposed Rule”). Pearson & Shaw design dietary supplements and health foods and license these formulas to carefully selected vendors and manufacturers. Commenters would be adversely affected by the Proposed Rule since it would substantially raise production costs and product sale prices, significantly reduce customer demand for dietary supplements, and has already reduced the number of manufacturers and sellers of their licensed formulations. Durk Pearson & Sandy Shaw Quality Control Experience Durk Pearson has patents in the area of oil-shale recovery, lasers, holography, and foods. He worked on manned aerospace programs from Gemini to the Shuttle and won numerous awards, including an award (attached as Exhibit 1) from the International Society for Testing and Failure Analysis (a professional organization) for his penetrating quality control and safety analyses. (In aerospace, quality control is a matter of life and death.) He wrote much of the safety manual for the Materials Processing Laboratory on the Shuttle. He had principal responsibility for determining whether prospective bidders on solid-rocket-engine carbon-composite throat packages for our strategic nuclear missile deterrent had production and quality assurance systems in place adequate to assure reliable operation of these ICBMs. Each bidder that qualified had its own systems and procedures; there were no ICBM rocket-engine throat CGMPs. (The throat of a solid-fuel rocket engine is the most highly stressed and most difficult to design and manufacture component, and the part most likely to fail catastrophically.) Durk took a triple major at MIT in physics, biology, and psychology, with a triple minor in electrical engineering, computer science, and chemistry, graduating from MIT in 1965 with a degree in physics. Sandy Shaw has a patent in foods. She received her degree from UCLA in chemistry in 1966 after taking a double major in chemistry and zoology and a minor in mathematics. Her work in the food industry has included supervision of quality control of all food production at a major canned food company. In the late 1960s, she designed and implemented the first computerized quality control system to be used at the company, doing all the computer programming herself. The Commenters include by reference the CGMP Economic Impact Assessment by Paul H. Rubin (hereinafter the Rubin Report, the Rubin Economic Analysis, or the Rubin Economic Impact Assessment) that was included with the joint comments of Essential Nutrition, Ltd., Life Enhancement Products, Inc., Durk Pearson and Sandy Shaw, Julian M. Whitaker, M.D., Advocare International, L.P., Vitamin Research Products, and American Nutrition Corporation, represented by Jonathan W. Emord, Claudia A. Lewis-Eng, Andrea G. Ferrenz, Kathryn E. Balmford, and Jonathan R. Goodman. In addition to the comments we have written here, we also include all those comments that are written directly upon the enclosed pages of our copy of the FDA’s Proposed Rule, adjacent to the portions of the Proposed Rule to which the comments pertain. Protecting the Public Health and Safety The FDA’s mission is to protect the public health and safety. The products that are actually brought to market are the instrument by which the public’s health and safety may be affected and, hence, are the proper object of the FDA’s attention. The products, not the factories, are in interstate commerce. Whether, as required under the proposed CGMPs, a particular piece of paper is or is not filed should not be considered evidence of adulteration or misbranding, as the FDA proposes, but of the failure to file a piece of paper. The products themselves entail all the necessary evidence of whether there is any adulteration or misbranding, such as by not including the ingredients stated on the label (misbranding), by not including the amounts of the ingredients as stated on the label (misbranding), or by including something that is dangerous to the public when used as recommended on the label (adulteration). Misbranding and adulteration are already illegal, and FDA already has all legal authority needed to seize such products and to punish those responsible. For further substantiation of this approach—looking at the actual standards met by the products themselves, rather than providing elaborate rules and regulations for the factories that made them—see Rubin’s analysis of “design standards” vs. “performance standards” in his Economic Impact Assessment, pp. 17–18. We Do Not Trust the FDA to Respect Our Constitutional Rights We are two scientists who earn much of our living by designing dietary supplements that we license to various companies that meet our standards. We have had to spend large sums of our own money, along with others who have also had to spend considerable amounts of money, to fight the FDA in the courts for violating our First Amendment rights to include truthful, nonmisleading scientific information about dietary supplements on their labels and in labeling. The courts repeatedly agreed with us that the FDA was violating (even willfully violating) our First Amendment rights. From the time (1994) that we filed a First Amendment suit to require FDA to allow a scientifically valid omega-3 DHA/EPA cardiovascular-disease health claim to the time that FDA finally obeyed the orders of the courts (2001), approximately 1,000,000 Americans died premature deaths that could have been prevented by use of this supplement. Hence, you may not wonder that we are concerned with the imposition of a system (as described in the Proposed Rule) that allows for immense and arbitrary discretion on the part of FDA inspectors. An example of FDA’s Proposed Rule is that a consultant must be qualified by training and experience to advise on his or her subject area. This, like many other of the rules, is obvious. However, the problem is that somehow we and the FDA may have different opinions on whether a consultant we hire is qualified by training or experience. If our products are not misbranded or adulterated, why should we have to share the FDA’s opinion of our consultant? Yet the FDA can brand our products adulterated if a consultant of whom they disapprove was associated with its manufacture or design, regardless of whether there is in fact any product defect. There is special danger in this Proposed Rule for those small companies, including the two of us, who have supported suits against the FDA on the basis of FDA’s violation of their First Amendment rights, as affirmed again and again by the courts. The 547 pages of rules and regulations set the stage for FDA agents to easily focus their attention on companies such as these, checking the myriad rules to find something they can use to punish them, such as an allegedly missing paper, or any aspect of factory design they can claim should have been done differently (such as “your washroom is too far from or too near to the mixer”). The Proposed Rule is peppered with many undefined terms (such as “adequate,” “sufficient,” and “qualified”); without definitions, it will always be possible for an FDA inspector to assert that something is not adequate, sufficient, or qualified. It is because the FDA has been unwilling to enforce the existing laws that already prohibit misbranded or adulterated products that this new scheme is hatched that adds a large new expensive and burdensome body of regulations to those the FDA has failed to enforce. We ask the FDA: why doesn’t the agency consistently enforce the existing laws against misbranded and adulterated products? With that failure as a record, why should we believe that the FDA, having a huge new body of regulations added to the ones it hasn’t enforced, will now (somehow) enforce the law against misbranded and adulterated products? From Rubin’s Report, it is clear that the costs of these new rules and regulations are far, far greater than the benefits. The Proposed Rule Includes FDA Fraud on the Public Furthermore, the FDA proposes to legalize its right to commit fraud on both consumers and companies by publicly branding a product as “adulterated” even if it is not defective in any way and contains exactly what it is labeled as containing. The public understands “adulterated” as meaning diluted (subpotent) or contaminated with harmful substances, such as rat feces or poison. 999,999 persons out of a million will not understand that an “adulterated” product can be faultless, but that the bureaucratic procedure is allegedly missing some obscure and unnecessary piece of paper. The government does not have the right to commit willful fraud! Moreover, the primary principle of American justice is that a person or entity is considered innocent until proven guilty, and in DSHEA Congress expressly created a statutory presumption of safety. The FDA, however, brands somebody’s product as adulterated or misbranded (having a defect as noted above), regardless of whether it actually has such a defect, because of the failure of the manufacturer to (for example) file a certain paper for the FDA inspectors. Not only is such a system of presumed guilt an unjust and unconstitutional one (and in violation of DSHEA), it gives the FDA immense powers to seize products that are in fact perfectly legal and to punish manufacturers who made those perfectly legal products but, because of a missing paper, are subject to FDA punishment. Is this the way to protect the public health? This proposal effectively controls every aspect of factory activity, establishing a regime that is easily abused, with huge costs but little benefit, while destroying our liberty interests in productive enterprise—over the innovative dietary supplement industry as well as over the free choice of consumers who wish to buy from it. As we note above, the protection of the public health requires that FDA regulate actual products that enter into interstate commerce. We do not see why the FDA should design plants, quality control systems, and computer systems. The FDA is unwilling to enforce current law to remove adulterated or misbranded products from the market, yet we are to believe that this qualifies the agency to exercise broad new powers to regulate nearly everything to do with the production of products and that they will not only enforce all this but that there will be improved health and safety without inordinate costs! We do not trust the FDA with these broad new powers. Certificates of Analysis The FDA has made it clear that dietary supplement companies cannot rely on the certificate of analysis on an incoming ingredient, regardless of how reputable the manufacturer (e.g., Roche Vitamins, Ajinomoto, Tanabe) is or how long it has been in business without problems with its ingredients (e.g., Roche Vitamins, Ajinomoto, Tanabe). This is a disincentive for dietary supplement companies to pay the premium prices that are usually charged for dietary ingredients by the most reputable companies. Rather than pay a premium price for the ingredient and then have to pay the additional costs for testing, many companies will likely choose to buy cheapo dietary supplement ingredients such as those from unidentified manufacturers in China that will probably pass all the necessary tests but are still more likely to contain contaminants undetected by tests than those from the highly reputable firms. A good example is the contaminated L-tryptophan from Showa Denko; there were no such toxic contaminants in the tryptophan manufactured by Ajinomoto or Tanabe. Showa Denko became a dominant source of L-tryptophan because they charged significantly less for their L-tryptophan. Their contaminated L-tryptophan met USP analytical standards (the contaminant was present at parts per million levels) and would not have been either prevented or detected by the CGMPs proposed by the FDA. To this day, FDA claims that they do not know the chemical identity of the contaminant or contaminants, which is the sole basis for FDA’s continuing prohibition on tryptophan importation. The toxin or toxins was an order of magnitude more toxic than VX nerve gas, the most potent of those that have been weaponized. One cannot test for an unknown toxin at these very low levels. We suggest, therefore, that the FDA allow reliance upon certificates of analysis on products from reputable firms such as Roche Vitamins, Ajinomoto, and Tanabe to provide authoritative analytical information on incoming ingredients in place of redundant, money-wasting extra testing that is likely to drive many companies to save money by using cheaper, potentially less safe ingredients. Do you (the FDA) really believe that you can enforce effective CGMPs in a Communist dictatorship like China? The Proposed Rule Is Not Modeled After Food CGMPs FDA cites (p. 44) its legal authority for the Proposed Rule. Section 402(g) of the Public Health Service Act gives the FDA explicit authority to issue a rule regulating conditions for manufacturing, packaging, and holding dietary supplements. Section 402(g)(2) of the act states that any such regulations “shall be modeled after current good manufacturing practice regulations for food and may not impose standards for which there is no current and generally available analytical methodology.” However, as we explain below, there is extensive evidence within the Proposed Rule that the CGMPs proposed are druglike rather than foodlike. First, FDA argues (p. 45) that when Congress used the term “modeled after,” Congress intended that the FDA use the food CGMPs as a “preliminary pattern” for the dietary supplement CGMPs. Regardless whether “preliminary pattern” is really what Congress meant by “modeled after,” it is clear that Congress did not intend that the CGMPs for dietary supplements be modeled after drug CGMPs. The FDA notes that if Congress had intended for the agency to adopt food CGMPs as the CGMPs for dietary supplements, they could have explicitly stated that. Instead, Congress specified that the dietary supplement CGMPs be “modeled after” food CGMPs, allowing for some modification. It is unlikely, however, that Congress would have specified that dietary supplement CGMPs be “modeled after” those for foods if they wished those for dietary supplements to be more onerous than those for foods, and, clearly, dietary supplement CGMPs were not to be modeled after drug CGMPs. On p. 46, FDA states that “dietary supplements can be considered as falling somewhere along the continuum between conventional food on the one hand and drugs on the other.” The Dietary Supplement Health and Education Act, however, made it clear that dietary supplements were to be considered and regulated as food. FDA pursues this drug analysis (pp. 46–47) by claiming that dietary supplements, unlike conventional food, contain ingredients that are consumed in very small quantities. That is not true. Spices and flavoring herbs are frequently used in foods in very small quantities, as are fortification vitamins (such as folic acid and niacin in flour and vitamin D in milk). FDA then claims that dietary supplements, unlike conventional foods, may be intended to have a specific physiological response. However, conventional foods are full of ingredients that have specific physiological effects (there would be no point in eating if they didn’t). For example, the choline in eggs and soybeans is a precursor to the neurotransmitter acetylcholine, required for memory and focus as well as muscular contraction. The omega-3 fatty acids in fish have important anti-inflammatory, anti-abnormal-clotting, and anti-arrhythmic effects. The American Heart Association’s Step 1 and Step 2 diets have been designed to have specific beneficial physiological and biochemical effects (such as lowering LDL-cholesterol) on patients with extant cardiovascular disease. On p. 51, the FDA states that “We note that certain terms Congress used in section 402(g)(2) of the act, i.e., ‘standards’ and ‘current and generally available analytical methodology,’ show that Congress intended to give us the authority to establish regulations in this rule that do not have parallel provisions in other food CGMPs.” Expressions such as “standards” and “current and generally available analytical methodology” apply to food science, so we do not see how these terms give FDA any authority to go beyond modeling their CGMPs after those of food. On p. 67, FDA notes that dietary supplement companies are not permitted, under the Proposed Rule, to use a supplier’s certification even though food companies may. On p. 71, FDA notes that food CGMPs do not have record-keeping requirements, but that the FDA doesn’t see that this prevents them from imposing those upon dietary supplements, as they do with drugs. This, however, is not modeled after food CGMPs. The FDA explains that many dietary supplements contain pharmacologically active substances. However, so do foods. They further explain that some dietary supplements may contain potential allergens. However, so do far more foods. On p. 244, the FDA commands: “Quarantine packaging and labels until your quality control unit tests or examines a representative sample to determine that specifications are met.” [Emphasis added] This is just one example in the Proposed Rule where FDA uses the same language, nearly word for word, as in the drug CGMPs. The FDA didn’t merely model these CGMPs after the drug CGMPs, they copied much of the drug CGMPs. On p. 330, the FDA makes estimates of record-keeping expense burden “based on our institutional experience with CGMP requirements for drugs.” On p. 414, the FDA notes that they “used a study of a medical device CGMP regulation to estimate the costs of record keeping.” Why would the FDA use their experience with drug CGMPs and a study of medical-device CGMP regulation to estimate the costs of record keeping in a CGMP modeled after food? Because the CGMP in the Proposed Rule is not modeled after food. Compounding Pharmacists and the Need for CGMPs Compounding pharmacists are small-scale manufacturers of drugs who are not required to follow the drug CGMPs and who make compounds that, if there is too much or too little of certain ingredients, or the wrong ingredients, can result in serious illness or even death. While compounding pharmacists work with potentially toxic materials in their practice, they have an excellent health and safety record, probably better than that of large companies that follow drug CGMPs. This seriously brings into question the idea that drug CGMPs and the huge costs they impose on companies (and the much higher prices for drug products that result) are actually improving the public health and safety. Moreover, compounding pharmacists, since they are not required to follow drug CGMPs, are very unlikely to be required to follow dietary supplement CGMPs. Hence we will likely see compounding pharmacists making individualized dietary supplements without being subject to CGMPs in the Proposed Rule. This parallel system, wherein some small manufacturers have to follow CGMPs and others do not, brings into question the requirement for FDA’s CGMPs to protect the public health and safety. The Proposed Rule Does Not Meet the Requirements of Executive Order 12866 Executive Order 12866 (originally issued by President Clinton) says agencies can “adopt a regulation only upon a reasoned determination that the benefits of the intended regulation justify its costs.” The Rubin CGMP Economic Impact Assessment finds upon analysis of the Proposed Rule that “benefits are no more than $13.9 million and costs are at least $860 million.” The alleged—not proven—benefits cannot justify the 61-times greater costs. We suggest that the FDA consider mandating performance standards rather than design standards. As explained in Paul H. Rubin’s CGMP Economic Impact Assessment (pp. 17–18), “Design standards mandate the actual procedure a regulated entity must follow to meet the regulation. Performance standards mandate the outcome, but allow the firm to choose the method for achieving the outcome. It is generally agreed that, where possible, performance standards are preferable since they allow the regulated firm more discretionary authority to achieve the desired outcome at lower cost. The standards in the Proposed Rule (and all alternatives considered on pp. 12221–12223) are strictly design standards: they mandate actual procedures that must be followed. “The FDA would do well to consider restructuring its mandate in terms of performance standards, such as no more than a specified level of contamination. Of course, given the low level of harm identified in the industry, it is highly likely that it would already satisfy any reasonable set of performance standards, indicating that regulation would serve no useful purpose.” If the FDA is bound and determined that it must impose a regulatory regime upon the industry, despite the industry’s low level of harm (when compared to foods in common form, to say nothing of drugs), then please do the least collateral damage to the industry and to consumers buying its currently remarkably safe and relatively inexpensive products by mandating performance standards, not design standards. Is it because the FDA is unable to define the desired outcome that FDA is, instead, attempting to mandate in its place a series of step-by-step instructions that achieve very small benefits for very great costs? Specific Comments On FDA’s Table 8 (pp. 375–376) This summary of health effects associated with dietary supplement recalls from 1990–1999 does not support FDA’s demand for CGMPs; on the contrary, it suggests that a vastly smaller and less expensive suite of tests would prevent all alleged harms that could have been prevented by FDA’s druglike CGMPs. Hypervitaminosis A: The health cost per event of $936 suggests that this did not cause serious harm. Even this small harm could have been prevented merely by analyzing the finished product for vitamin A content. Very expensive CGMPs are not needed. Hypervitaminosis D: The health cost per event of $1,022 suggests that this did not cause serious harm. Even this small harm could have been prevented merely by analyzing the finished product for vitamin D content. Very expensive CGMPs are not needed. Hypervitaminosis B6: The health cost per event of $8,868 suggests that this did not cause serious harm. Even this small harm could have been prevented merely by analyzing the finished product for vitamin B6 content. Very expensive CGMPs are not needed. Hypervitaminosis B3: The health cost per event of $4,258 suggests that this did not cause serious harm. Indeed, the average healthy adult can safely consume 800 mg of niacin per day without harm (though with potentially alarming brief flushing and pruritus). (A single capsule of this product made it very clear to the customer that the product was defective; there was no opportunity for liver damage to occur.) Even this small harm could have been prevented merely by analyzing the finished product for niacin content. Very expensive CGMPs are not needed. Hypermineralosis Zn: The health cost per event of $4,228 suggests that this did not cause serious harm. Even this small harm could have been prevented merely by analyzing the finished product for zinc content. Very expensive CGMPs are not needed. Hypermineralosis Cu: The health cost per event of $369 suggests that this did not cause serious harm. Even this small harm could have been prevented merely by analyzing the finished product for copper content. Very expensive CGMPs are not needed. Hypermineralosis Fe: The health cost per event of $374 suggests that this did not cause serious harm. Even this small harm could have been prevented merely by analyzing the finished product for iron content. Very expensive CGMPs are not needed. Hypermineralosis F: The health cost per event of $938 suggests that this did not cause serious harm. Even this small harm could have been prevented merely by analyzing the finished product for fluoride content. Very expensive CGMPs are not needed. Hypermineralosis Se: The vastly excessive selenium caused death, which is certainly serious harm. This great harm could have been prevented merely by analyzing the finished product for selenium content. Very expensive CGMPs are not needed. The above nine recalls and the harms associated with them could have been entirely prevented by simple and relatively inexpensive quantitative analysis of the finished products for all ingredients that have the potential for significant harm if the capsule or pill is comprised entirely of that ingredient. Very expensive CGMPs are not needed to prevent these types of harms. Salmonella and Klebsiella pneumoniae are not problems in most dietary supplements, though aqueous herbal extracts and other nondry products are subject to bacterial contamination, just like foods. The requirement for food CGMPs (such as bacteriological testing of the finished products) for these types of products would have prevented these harms. Very expensive druglike CGMPs are not needed. Glass fragments in products can be prevented by use of appropriate food CGMPs whenever glass is present. Very expensive druglike CGMPs are not needed. Yellow #5 (undeclared), Yellow #6 (undeclared), Red #40 (undeclared), Blue #2 (undeclared), lactose (undeclared), sulfites (undeclared), and ephedra (undeclared): What do these incidents all have in common? The harms were caused by the fact that ingredients were not declared, a practice that is already illegal as both misbranding and adulteration. These ingredients were purposely added and not listed on the labels. Very expensive druglike CGMPs are not needed; FDA should simply enforce existing laws against misbranding and adulteration. Digitalis: The medical plant containing digitalis contaminated the herb plantain, with deadly results. Since the supplement manufacturers received the plantain in powdered form, the contamination could not have been detected by visual or microscopic inspection, even by a “properly trained plant taxonomist.” It is economically unfeasible to test every batch of herbs for the thousands of known plant toxins (at about $200 per test for each toxin). Druglike CGMPs at the plant-material provider would arguably have reduced the risk of this contamination, but since the provider was almost certainly located in China, Korea, Russia, or some other Second or Third World foreign country, it is very unlikely that FDA’s CGMPs could have been enforced there. We find it very interesting to note that, according to our medical journal accounts of this incident, FDA knew that over 100 supplement manufacturers were using this contaminated batch of plantain, but apparently did not bother to contact most of them! Gross dereliction of duty by some FDA personnel may be a dangerous problem that will not be corrected by imposing CGMPs on the supplement industry. Very expensive druglike CGMPs at the supplement manufacturers would not have prevented this incident, and very expensive druglike CGMPs could not have been verifiably imposed on a foreign herb vendor. Eosinophilia-myalgia syndrome: As the U.S. Centers for Disease Control and Prevention has emphatically stated, the EMS epidemic was caused by contaminated tryptophan produced by one foreign ingredient manufacturer (Showa Denko). To this day, FDA prohibits any importation of tryptophan from any manufacturer (including Ajinomoto and Tanabe, whose tryptophan was never contaminated), on the basis that the toxic contaminants have never been identified. Tryptophan is not manufactured domestically. (These contaminants were ultratoxins; they were present at parts per million levels and were more toxic than VX nerve gas. The contaminated tryptophan met USP standards for purity.) Since the toxic contaminants were unknown at the time of the incident and were of unprecedented toxicity, neither FDA’s proposed CGMPs at the manufacturer nor CGMPs at the dietary supplement manufacturers who used the contaminated tryptophan could have prevented this tragedy; hence, as Dr. Paul Rubin said, this incident cannot legitimately be used to justify the imposition of CGMPs. No contaminated tryptophan was used by any of our licensees in manufacturing our tryptophan-containing formulations; we permitted only Ajinomoto and Tanabe tryptophan to be used, because of their very extensive experience and long history of safely manufacturing this substance—which, in our opinion, is much more important than FDA’s proposed CGMPs.