The Safety of Dietary Supplements and Pharmaceuticals

Introduction

The Dangers of Prescription Drugs

What is a Dietary Supplement?

The Alleged Dangers of Dietary Supplements

Dangerous Dietary Supplements and Safe Drugs

Dietary Supplements Regulatory Reforms?

Conclusion

Links and References

The dangers of prescription drug treatments are well known and have been repeatedly confirmed by scientific studies conducted in various countries around the world ( see Holistic Medicine Sets the Standard for Safety, Science or Progress? ). Extensive studies of hospital patients in the U.S. have revealed that adverse drug reactions are responsible for an absolutely staggering and disgraceful 100,000 deaths annually ( 1, 2, 3, 4 ). Furthermore, a massive 2 million U.S. hospital patients also experienced serious but non-fatal reactions to medical d rugs ( 1, 3, 4 ). These figures, which do not include adverse drug reactions occurring outside hospitals, are also typical of figures from other countries such as Australia ( see Holistic Medicine Sets the Standard for Safety ).

The point must be emphasised that these figures have been repeatedly substantiated. Their reliability is beyond question since they have been derived from detailed hospital data and they have been compiled by various medical experts.

It must also be realised that, by and large, in spite of these indisputable facts, the risks of medical drug treatment are considered quite acceptable and are generally not questioned. The fact that these risks are not confined to one or two drugs but are spread across such a broad range of medical drugs has clear implications for the use of all drugs. Nevertheless, the future of medical drugs never seems to be questioned, especially by those who are most familiar with their hazards. Mainstream medicine has made the judgment that 100,000 deaths and 2 million serious adverse reactions in the U.S. alone every year is insufficient to question the principles of drug therapy. The benefits they repeatedly say, outweigh the risks.

Let us now consider dietary supplements.

The term "dietary supplement" is most appropriately described as a legal term rather than a literally accurate term. Most people would probably consider that a dietary supplement is a substance which may be taken as a supplement but which is also contained in a normal diet. This is not the case however since this term is often used to describe all those over the counter substances for which, unlike drugs, no specific therapeutic purpose has been generally accepted by modern medicine. In other words, dietary supplements, as legally defined, are not necessarily derived from normal dietary constituents at all. Additionally, substances defined as dietary supplements in the U.S. are not necessarily dietary supplements in other countries.

One example of the confusion between dietary supplements and drugs is a substance termed gamma-hydroxybutyrate or GHB and its precursor products ( 5, 6 ). This substance, the use of which as an anaesthetic drug was apparently abandoned in Europe in the 1960's because of its ill effects ( 5 ), subsequently became a dietary supplement in the U.S.A. (5 ). Even when GHB was banned as a dietary supplement suppliers simply reverted to selling GHB precursors which had the same ill effects (5 ).

Other substances sold as dietary supplements include hormones and herbal medicines. Various hormones which have no prescribed application in mainstream medicine may be regarded as dietary supplements in some countries although they may be restricted in other countries. Herbs too are regarded as dietary supplements although a clear distinction needs to be made between non-toxic nutritive herbs and the more toxic and pharmacologically active herbs. While many herbs have a gentle nourishing restorative effect there are various others which more closely resemble synthetic drugs due to their toxicity and their potent pharmacological effects. One of these is the Chinese herb Ma Huang or Ephedra ( 7, 13 ), a herb which is commonly blamed for causing significant ill effects and which is therefore banned or restricted in some countries.

The term "dietary supplement" is clearly a very vague term which is used to loosely describe many over the counter medicines. If such a general classification is to be employed then it would seem sensible that there should also be sub-categories based upon the nature and toxicity of each individual substance.

In order to assess the validity of claims about the dangers of dietary supplements I have selected a recent study by Palmer and colleagues entitled "Adverse Events Associated with Dietary Supplements ( 5 ) as being representative of recent criticisms. Since this study has been cited in the media as providing proof of the dangers of dietary supplements it is essential that it has a sound scientific basis and is free of bias. Since Palmer and colleagues concluded ( 5 ), as a result of their study, that "research into hazards and risks of dietary supplements should be a priority", and there should be "mandatory reporting of adverse events", it is vitally important that the basis of this report is examined in some detail.

The first point that should be made regarding the study of Palmer and colleagues is that, unlike claims about the dangers of prescription drugs which are based upon detailed hospital data (1, 2), their claims about dietary supplements are based initially upon telephone calls to poison control centres (5 ). To quote Palmer and colleagues ( 5 ): "in 1998, 11 poison control centres in the USA recorded details of 2332 telephone calls about 1466 ingestions of dietary supplements, in 784 of which patients had symptoms. We used a multitiered review process (kappa 0-42) to select 489 cases for whom we were at least 50% certain that their negative events were associated with dietary supplements."

Stated another way, the 489 cases which form the focus of the Palmer report were cases for whom the researchers were up to 49% doubtful that the negative events reported were even "associated" with dietary supplements. According to this analysis it would seem that there is little evidence here that would meet the normally accepted criteria for scientific evidence.

In defence of their decision to base their study upon data derived from poison control centres Palmer and colleagues claim that ( 5 ) "adverse events associated with dietary supplements are recorded only through voluntary reporting" and they "do not have to undergo the mandatory prelaunch testing and postlaunch surveillance required of prescription drugs." In spite of these claims however, there is an attitude of "silence" within mainstream medicine since doctors commonly prefer not to report adverse reactions to the drugs they prescribe ( see Holistic Medicine Sets the Standard for Safety ). The situation with toxic prescription drugs is therefore much more serious than it is for dietary supplements.

The point must be emphasised that the detailed hospital data which is available for assessing the dangers of prescribed drugs owes its existence to the simple fact that serious reactions to drugs result in the need for hospital treatment, even if the drug was not administered in a hospital. In Australia alone some 80,000 patients are admitted to hospitals every year because of reactions to prescribed drugs ( see Holistic Medicine Sets the Standard for Safety ). Hospital data for adverse drug reactions it should be noted, is not the result of some automatic reporting system or an elaborate surveillance system of drug reactions, rather it is compiled from hospital data by conscientious researchers. Drug reactions and iatrogenic diseases also tend to be given an extremely low visibility in official statistics. Death certificates also tend to avoid mentioning contributory drug reactions.

Why is it that this type of detailed hospital data is not available for dietary supplements? Is it because so few people are admitted to hospital as a result of the use of these products?  The point made by Palmer and colleagues ( 5 ) that "voluntary reporting" is an "inherently weak form of surveillance," underlines the importance of a reliable source of data.

Another significant problem with data from poison control centres is the large number of incidents which result from intentional abuse, misuse, or attempted suicide. According to the 2001 AAPCC report ( 9 ), of 1074 fatalities reported in that year, 74% of adolescent deaths and 79% of adult deaths were described as "intentional". Of the total 1074 fatalities, 51% were attributed to suicide. Similarly, according to Deng and co-workers ( 10 ) in regard to 273 poisonings allegedly due to traditional Chinese medicine, "all occurred because of suicide attempts, accidents, or erroneous or improper use of processing." The report of Palmer and colleagues ( 5 ) was based upon "2332 (telephone) calls, reporting 1466 ingestions, 36% (534) of which were  unintentional". Palmer continues: "half the exposures produced signs or symptoms (741), and reviewers decided that 66% (489) of these were related to ingestion of dietary supplements."

Although Palmer and Colleagues (5 ) claim that 64% of the total 1466 ingestions were "intentional" they fail to specify or quantify the various categories of intentional use such as how many were due to misuse, abuse or attempted suicide. This is in spite of the fact that these workers claim that ( 5 ) "another strength of our study was the inclusion of intent of use." While failing to provide any precise figures, even though such information is a standard part of the TESS ( Toxic Exposure Surveillance System ) data utilised by poison control centres ( 9 ), Palmer and colleagues simply point out, with regard to attempted suicides, "the rates were close between dietary supplements and other substances." In spite of the fact that the basic TESS data utilised by the AAPCC specifies different categories of intentional ingestions ( 9 ) including "intentional misuse", "intentional abuse", and "suspected suicidal", these distinctions were omitted from the Palmer report ( 5 ). Given the extremely high incidence of intentional, abuse, misuse, or attempted suicides amongst reports to poison control centres ( 9 ) the omission of this data from the Palmer report makes accurate interpretation of the data impossible.

Not only did Palmer and colleagues have access to the standard TESS data, but furthermore, as they emphasise ( 5 ), they compiled an additional questionnaire to provide more information about causality and reason for ingestion. Their failure to quantify the various categories of misuse, abuse, or overdosage invalidates much of the data and raises serious questions about the credibility of the study.

If we are to accurately assess the dangers of dietary supplements it is imperative that each individual case of alleged adverse reactions is analysed in some detail, particularly in regard to other possible contributory factors such as ingestion of other medicines or drugs. Although Palmer and colleagues supply some general data about ingested products which contained a number of different ingredients, no breakdown of the data is provided to identify those cases which involved concurrent ingestion of other medicines or drugs. This is a fundamental requirement if there is to be any ability to accurately determine the cause of the adverse reaction. Perhaps users of prescription drugs or other medicines were excluded from the study, however this was not stated in the report.

Instead of listing adverse reactions on a case by case basis which would permit analysis of other contributory factors, as is done in the AAPCC report, in Appendix 1 of the Palmer report ( 8 ) alleged adverse reactions are grouped according to the nature of the symptoms while in Appendix 2 ( 7 ) reactions are grouped according to the dietary supplement with which they are allegedly associated. No attempt is made to individualise the data so that intent of use and other contributory factors may be readily identified in each case. In other words, a person who is claimed by Palmer and colleagues to have experienced an adverse reaction to a dietary supplement may also have consumed large quantities of alcohol and may have been ingesting 20 prescribed drugs daily. They may also have taken a huge overdosage of the particular dietary supplement which has been implicated in an attempt to commit suicide. The data provided by Palmer and colleagues make identification of such individual factors impossible. This is in spite of the fact that the poison control centre data used by Palmer and colleagues normally lists this type of information.

Since Palmer and colleagues conclude ( 5 ) "that at least some dietary supplements have potential for substantial hazard", it is interesting to observe the data supplied by these workers regarding individual dietary supplements. According to Appendix 2 of the report ( 7 ), Ma Huang or Ephedra was responsible for the largest number of adverse reactions, a total of 90 reports representing 80 multiple ingredient ingestions and 10 single ingredient ingestions, a result which is not too surprising ( 13 ). If we look at single ingredient ingestions the most frequently reported products were Melatonin with 31 reports and St John's Wort with 21. Even lecithin was said to be responsible for 21 multi-ingredient reports with symptoms ranging from agitation and dizziness to vomiting and oedema. Strangely, if we examine the list of symptoms attributed to lecithin in Appendix 2 ( 7 ), not one of these symptoms were attributed to lecithin in Appendix 1 ( 8 ). The mineral chromium was said to be responsible for a total of 60 reports of vomiting while zinc caused a total of 34 reports of symptoms ranging from anaemia to vomiting and skin rashes.

In spite of claims about the dangerous nature of dietary supplements, according to the Inspector General of the U.S. Department of Health and Human Services ( 14 ): "currently about 60% of Americans take some form of dietary supplement everyday without any apparent problems." Similarly, according to U.S. Congress ( 16 ): "dietary supplements are safe within a broad range of intake, and safety problems with the supplements are relatively rare."

As I have indicated previously, the dangers of prescription drugs have been repeatedly scientifically validated by extensive hospital data. However, the poison control centre data used by Palmer and colleagues (5 ) to assess the alleged dangers of dietary supplements may also be used to assess the dangers of pharmaceutical substances (even though this data was omitted from the Palmer report). By using the same source of data a more direct comparison can be made regarding the risks of both dietary supplements and pharmaceutical products although this data does have definite limitations and does not compare with much more detailed hospital data. It is unfortunate that Palmer and colleagues chose to focus exclusively upon dietary supplements in their report and thereby make any direct comparison with pharmaceutical products impossible.

There are two fundamental reasons why an accurate comparison with pharmaceutical products is absolutely essential. Firstly of course, evidence cannot be considered scientific or valid if it is characterised by bias. Since it must be assumed that studies into the safety of therapeutic substances are motivated by concern about the safety of consumers, such concerns should not be seen to be conveniently confined to certain types of products while there is no concern about other products. Concerns which are not seen to be proportional to evidence of known risk, are not consistent with the interests of consumers or with public health. Secondly, since the level of safety which has been achieved by modern medicine generally seems to be regarded as being quite acceptable and is the result of the "mandatory prelaunch testing and postlaunch surveillance required of prescription drugs" ( 5 ), those same standards of safety should also be acceptable for dietary supplements. In fact, since the safety standards achieved by modern medicine seem to be held up as the standards to which others should aspire also, it is necessary to evaluate the probable effects of these standards if they are applied to dietary supplements as has been suggested ( 5 ).

Let us use the 2001 Annual Report of the AAPCC (9) to make a direct comparison between pharmaceutical products and dietary supplements.

According to the 2001 AAPCC report ( 9 ) a total of 9 deaths out of the total 1,074 deaths in that year were attributed to dietary supplements. Of these, 3 were due to adverse reactions, 2 to therapeutic errors, and one to suicide, while the reasons for the remaining 3 fatalities were unknown. Five of these 9 fatalities were associated with ingestion of Ma Huang, either alone or in combination with drugs. A further two cases were said to be caused by dinitrophenol, a highly toxic chemical, which although banned in 1938 is still used by some people for weight loss ( 11, 12 ).

On the other hand, of the substances implicated in fatal poisonings in 2001, 84.6% were pharmaceuticals (9 ), with analgesics being implicated as the primary cause of death in 32% of fatalities or 341 deaths (9 ). This compares with 0.8% for all dietary supplements combined ( 9 ), even including substances such as dinitrophenol. Interestingly,  the anti-asthma drug theophylline alone was responsible for 15 deaths, 66% more than all the available dietary supplements combined.

The claim by Palmer and colleagues that "adverse events related to supplements" had "outcomes no less severe than those for all other product exposures" is not consistent with the fact that, according to AAPCC data, by far the majority of severe or fatal reactions reported to poison control centres in 2001 were due to pharmaceuticals. According to the 2001 AAPCC report ( 9 ): "pharmaceutical exposures also had more severe outcomes. Of substances implicated in fatal cases, 84.6% were pharmaceuticals, compared with only 46.6%, in nonfatal cases. Similarly, 83.5% of substances implicated in major outcomes were pharmaceuticals." These clear facts seem to conflict sharply with the analysis of Palmer and colleagues ( 5 ): "the frequency distribution of outcomes associated with dietary supplements seemed no less severe than those from other calls about events not related to dietary supplements."

It is claimed that much more stringent regulations are required to control the marketing of all dietary supplements ( 5, 13, 14, 15 ) although such proposed changes often seem to be justified by reference to one or two of the more toxic products (13 ). Central to any such proposal is the belief that the current situation in the pharmaceutical industry is the result of established standards of safety, standards with which dietary supplements should also comply.

Although Palmer and colleagues omit from their study data which would make possible a direct comparison between the dangers of prescription drugs and dietary supplements, they nevertheless propose that the type of "monitoring" and "surveillance" which is applied to prescription drugs should also be applied to dietary supplements. Such suggestions seem to be based upon an assumption that prescription drugs have attained a totally acceptable level of safety. These workers claim that the "voluntary reporting" applied to dietary supplements is an "inherently weak form of surveillance" and what is therefore required is "mandatory reporting for adverse events related to dietary supplements." Palmer and colleagues continue: "as a minimum, the difficulties we have described suggest the need for a comprehensive register of dietary supplements, strengthened surveillance (particularly mandatory reporting of adverse events ), enforceable definitions of what constitutes a risk to safety, and what circumstances warrant product recall."

The reader is reminded at this point that the situation in mainstream medicine regarding adverse drug reactions and iatrogenic illnesses is characterised by a code of silence whereby doctors refuse to report such events ( see Holistic Medicine Sets the Standard for Safety ).

The actual safety record that has been achieved in real life by the pharmaceutical industry contrasts dramatically with claims that "drug products must be proven safe and effective for their intended use before marketing" ( 15 ). Of course scientific evidence over the past century has repeatedly demonstrated that prescribed drugs are neither safe nor effective ( see But What About Efficacy?, Holistic Medicine Sets the Standard for Safety, Health Trends, Science or Progress? ). The safety of prescription drugs is such that they have become one of the leading causes of death throughout the world ( see Holistic Medicine Sets the Standard for Safety, Health Trends, Science or Progress?  ). This is a simple scientific fact. As far as efficacy is concerned drugs have also failed miserably. Drug treated diseases such as asthma, diabetes, arthritis, heart disease, and mental diseases are all increasing to epidemic proportions but yet doctors and health authorities continue to claim that the drugs used to treat these diseases have been proven to be effective. Do they mean by "effective" that these drugs will do absolutely nothing to arrest the underlying cause and will therefore permit these diseases to flourish? According to abundant scientific evidence pharmaceutical drugs have been repeatedly shown to be unsafe and ineffective and if this is the result of the current regulatory system then it most certainly should not be applied to dietary supplements. Claims that pharmaceutical manufacturers are required by law to report adverse effects ( 5, 14, 15 ) should be seen in the light of the actual results that that have been achieved in the real world.

In their eagerness to apply current pharmaceutical regulations to dietary supplements some authorities seek to blur the distinction between nutrients and drugs. In fact, it seems that the fundamental mechanism underlying the biological effects of nutrients are very poorly understood by reductionist medical science. Suggestions that any substance which has "biological activity" is in fact a drug and therefore must, by definition, be dangerous ( 13 ), reveal very little understanding of the constructive functions of nutrients ( see Nutrition and Megavitamins ). If metabolism has been deranged in some way because of a nutritional deficiency then of course the appropriate nutrient will have "biological activity" but does it therefore follow that once the deficiency is corrected the nutrient will continue to have biological activity and hence become toxic? This is the nature of drugs not nutrients.

According to Fontanarosa and co-workers ( 13 ), any substance which has "biological activity" and may cause "physiologic changes" must be dangerous: 

In spite of the claims by Fontanarosa and co-workers ( 13 ), few would suggest vitamin B1 does not have "biological activity" when administered to those suffering from a vitamin B1 deficiency, but in spite of this, most experts would readily acknowledge that this vitamin will have no biological effect whatsoever if it is administered to a person who has adequate vitamin B1 status. The same is true of folic acid when it is employed to prevent birth defects and vitamin B12 for pernicious anaemia. What about the use of protein to reverse protein calorie malnutrition? Does this mean dietary protein is toxic? Even conditions such as heart disease and cancer may be prevented by diet but does this mean that food is toxic? As I have indicated elsewhere ( see Nutrition and Megavitamins ) this same reasoning also applies to many herbs since often herbs depend upon their nutritive qualities for their various effects. It seems that those who are intricately involved in drug treatment are incapable of understanding the constructive effects of nutrients. To them it seems, "nourishment" may be obtained by either nutrients or drugs since there is no distinction.

Measures that will improve the safety of consumers must be applauded, however, such measures must be fair and based upon scientific evidence. Voluminous scientific evidence reveals that the safety record of the prescription drug industry is absolutely disgraceful and if we are concerned about safety this is where our first priority should lie. Suggestions that dietary supplements would "benefit" from the same "monitoring" or "surveillance" which is applied to the pharmaceutical industry are inconsistent with the available scientific evidence. In fact the enviable safety record which has already been achieved by alternative medicine has set a bench mark that all those in the medical and pharmaceutical industries should aspire to.

History clearly shows that the reductionist approach of modern medicine, whereby safe herbs are concentrated and purified to form toxic drugs, is an inherently dangerous approach to health care. The highly profitable products produced in this way have repeatedly been shown to be highly dangerous and generally much more dangerous than the original herbs from which they were produced. We have learned, through human suffering, that this is most certainly not the way to go if our primary concern is safety. It would be an absolute catastrophe if medicine's current system, if applied to all dietary supplements, produced the same tragic results as have been achieved by the pharmaceutical industry.

In spite of the fact that the study of Palmer and colleagues has failed to accurately establish the dangerous nature of dietary supplements, there is no doubt that some substances which presently fall into this category are indeed dangerous. Toxic substances such as GHB and dinitrophenol more closely resemble drugs than dietary supplements. Various hormonal substances which are classified as dietary supplements in some countries are also capable of causing significant ill effects if used inappropriately. The stronger more toxic herbs, such as Ma Huang, Aconite and Pokeroot, are also capable of causing very serious ill effects. However, since the total number of "highly toxic" or "dangerous" dietary supplements which are commonly available probably amount to much less than 1% of the total market, there is absolutely no justification for new safety based regulations for the remaining 99% of safe products.

For recent changes to dietary supplement legislation in America see the FDA web site ( 17 ).

Conclusion

For the following reasons evidence of the dangerous nature of dietary supplements has not been clearly established by Palmer and colleagues ( 5 ).

• The researchers themselves admit there is up to 49% doubt as to whether the adverse events described in their study were even "associated" with dietary supplements.
• No attempt is made to identify or quantify adverse events which were the result of misuse, abuse, or attempted suicide.
• No attempt is made to exclude other possible contributory factors such as drugs and alcohol in those who experienced adverse events.

Not only are there serious questions about the reliability of the data regarding dietary supplements but the complete omission of data relating to safety of pharmaceuticals and prescription drugs could be interpreted by some as demonstrating a rather selective concern about the safety of therapeutic substances. It is odd that studies into the safety of therapeutic substances always seem to be confined either to drugs and pharmaceuticals on the one hand, or dietary supplements on the other, but never do these studies include both, and therefore apply the same standards of safety to both. Concerns about protecting the public from adverse reactions to therapeutic substances are commendable indeed, however, if these concerns are seen to be confined to some products and do not include other more dangerous products, then a most undesirable impression is created. It is vitally important that health authorities act in accord with the available scientific evidence and are seen to be unbiased.

There is tendency within mainstream medicine to suggest that if there is evidence that even one herb is dangerous then this somehow has implications for all other herbs and justifies general condemnation of all dietary supplements. Of course if this standard were applied to prescription drugs then all such drugs would have been banned long ago. What is needed here is a system of grading dietary supplements according to their toxicity and potential for causing ill effects. Herbal medicines are already graded in some herbal textbooks as being of low, moderate, or high toxicity, and this system should be extended, legitimised, and publicised. Suggestions that all dietary supplements should be controlled or monitored by mainstream medicine are clearly not supportable and not in the best interests of public health.

When it comes to alternative medicines such as dietary supplements, mainstream medicine has a very serious credibility problem. For more than 100 years mainstream medicine has seized any opportunity to denigrate alternative therapies and describe them as being useless or simply quackery. They have displayed an absolute determination to oppose publication and scientific research of alternative therapies. Attempts by those traditionally involved in alternative medicine to encourage scientific research have been constantly frustrated by the negative attitude of those involved in the medical and pharmaceutical industries. Now all of a sudden however, these alternative or "quackery" therapies are so potent that they should be controlled by mainstream medicine. We are expected to accept that modern medicine now wishes to control the very therapies they have been describing as quackery for more than 100 years.

According to a huge amount of evidence the medical and pharmaceutical industries are driven predominantly by self interest and profits rather than genuine public health concerns. In fact, science has become hijacked by commercial interests. Science today is frequently no more than a marketing tool and a means of justifying commercial decisions. It is difficult indeed to draw any other conclusion when faced with the bizarre nature and direction of medicine's attitude to alternative therapies. Why for instance does mainstream medicine have such a persistent and extreme reluctance to avoid applying the same safety standards to prescription drugs as they do to dietary supplements? Any researcher who expresses concerns about the safety of alternative therapies but who fails to mention the much greater hazards of prescription drugs is quite simply not expressing an attitude which is in the best interests of public health.

If we are to move forward in a logical and structured manner then what is needed is an unbiased study which includes both pharmaceutical substances as well as dietary supplements and which applies the same standards of safety to both. If safety is our primary concern then there will be no distinction. Distinctions are a product of self interest, not science or compassion. Corrective action could then be applied according to the areas of greatest risk. Research studies should not be biased unless towards safety and the public interest.

As far as public health is concerned the available evidence is abundantly clear. Since less than 1% of dietary supplements have been found to constitute any significant health hazard there is absolutely no justification for additional safety based regulatory controls of 99% of available dietary supplements. Pharmaceutical drugs on the other hand continue to cause a huge number of adverse reactions and fatalities. According to abundant scientific evidence therefore, urgent reforms are needed to improve the safety of pharmaceutical drugs and prevent adverse reactions. To this end there should be mandatory reporting of drug reactions in death certificates and specific inclusion in official statistics. There should also be severe penalties for doctors who fail to report significant adverse reactions. Given unbiased consideration of the evidence, no other conclusion is possible.

Links and References

1. http://jama.ama-assn.org/cgi/content/abstract/279/15/1200  See: Jason Lazarou, et al, JAMA. 1998;279:1200-1205.
2. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=Pub  See: Starfield B., JAMA 2000 Jul 26;284(4):483-5.
3. http://www.cnn.com/HEALTH/9804/14/drug.reaction/
4. http://www.mercola.com/1998/archive/hospital_drug_reactions.htm
5. http://pdf.thelancet.com/pdfdownload?uid=llan.361.9352.original_research.  See: Palmer, M.E., et al, Lancet, 361, 101, 2003.
6. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=Pub  See: Dyer JE., Am J Emerg Med 1991 Jul;9(4):321-4.
7. http://www2.landspitali.is/research/lancet/app2.htm  See also: Palmer, M.E., et al, (Append. 2), Ref No. 5.
8. http://www2.landspitali.is/research/lancet/app1.htm  See also: Palmer, M.E., et al, (Append. 1), Ref No. 5.
9. http://www.aapcc.org/Annual%20Reports/01report/2001%20TESS%  See: Litovitz, T.L., et al, 2001 Annual Report of the American Association of Poison control Centers Toxic Exposure Surveillance System, Amer. J. Emerg. Med., 20, 391, 2002.
10.http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=Pub  See: Deng JF, Lin TJ, Kao WF, Chen SS., Vet Hum Toxicol 1997 Apr;39(2):106-14.
11.http://www.jrnl.net/news/00/Jan/jrn120050100.html  See: Marian Burros, Low-carb diet makes another resurgence, New York Times News Service.
12.http://www.eco-usa.net/toxics/2nphenol.shtml See: Toxicological Profile for Dinitrophenol 1995, Agency for Toxic Substances and Disease Registry United States Public Health Service
13.http://jama.ama-assn.org/cgi/content/full/289.12.1568v1  See: Phil B. Fontanarosa, et al (editorial), JAMA. 2003;289:1568-1570.
14.http://oig.hhs.gov/oei/reports/oei-01-00-00180.pdf  See: Adverse Event Reporting for Dietary Supplements, Office of Inspector General, US Department of Health and Human Resources, April, 2001.
15.http://www.cfsan.fda.gov/%20~dms/ds-oview.html  See: Overview of Dietary Supplements, U. S. Food and Drug Administration, Center for Food Safety and Applied Nutrition, January 3, 2001.
16. Dietary Supplement Health and Education Act, 1994, Pub. L.No. 103.417 sec 2. Cited by ( 14 ).
17.http://www.cfsan.fda.gov/~dms/supplmnt.html