top of page
Various medical capsules and tablets in hexagonal jars in the form of honeycomb.jpg

Serious Adverse Events: Are SAEs as underreported as GAO suggests? (Part II)

In Part II of this two-part series, we pick up from Part I by further examining the numbers generated in the first article. The numbers reported in Part I represent dietary supplement exposure data derived from an initial examination of the American Association of Poison Control Centers (AAPCC) National Poison Data System (NPDS) Annual Reports. In this article, we will further refine these numbers and then evaluate their significance with regard to the dietary supplement products most commonly involved in “Adverse reaction” exposures[1] and the frequency and kinds of serious outcomes that are associated with adverse reactions to dietary supplement products. During the following discussion, it is important to remember that the report of a serious adverse event (SAE) received by the United States Food and Drug Administration (FDA) from a supplement manufacturer or a Poison Center (PC) does not signify causality – that is, the report indicates a serious health event occurred, but not that the dietary supplement caused or even contributed to the event.[2] Of the 6307 SAE reports (SAERs) FDA received from 2008-2011, 71% (4478) came from industry. (As explained in Part I, industry is required by law to submit all reports of SAEs that they receive to FDA.) The remainder, 1829 reports or about 457 reports per year, were submitted directly to FDA by consumers or others who did not contact industry. Earlier this year, the Government Accountability Office[3] (GAO) raised the concern that FDA may not be receiving information on all SAEs because consumers and others might instead be reporting these events to PCs. When the possibility was examined, PCs were found to have received over 1,000 more “Adverse reaction” calls related to dietary supplements (4863 total) than reports received by FDA (3827 total) for the same three-year period (2008-2010). Although these numbers were used only for direct comparison in the GAO report,[4] the larger number of calls to PCs was taken to indicate that consumers and others had contacted PCs without also providing reports to industry or FDA.

Considering the FDA reports first, in the 6307 SAERs submitted from 2008-2011, the following outcomes were reported: 3370 (53%) unspecified serious medical events, 1836 (29%) hospitalizations, 1272 (20%) serious illnesses or injuries, 512 (8%) life-threatening conditions, and 92 (2%) deaths.[5] FDA was able to establish “certain” causal relationships between reported SAEs and ingestion of the identified dietary supplements in only 212 (3%) reports, an average of 53 “certain” incidents per year. “Possible” or “probable” causal relationships were determined for 1884 (30%) reported events. For the remaining 4211 (67%) SAERs, the available information was not sufficient for a relationship to be established (as “possible,” “probable” or “certain”) or dismissed.[6] Interestingly, in no instance was a potential relationship between a dietary supplement and an SAE dismissed, which makes it appear that dismissal is not viewed as an option. In comparison, our initial examination of the AAPCC NPDS Annual Reports (Bronstein et al., 2009, 2010, 2011, 2012) over the four year period from 2008–2011 yielded an average of 4219 “Adverse reaction” exposure cases associated with dietary supplements (4055 in 2008, 4104 in 2009, 4295 in 2010, 4422 in 2011), which equates to about 0.17% of all confirmed human exposure cases. (The AAPCC logged and closed an average of 4,047,560 calls per year from 2008-2011; on average 2,422,308 (60%) were confirmed to be human exposures of various kinds. Of these, about 247,055 calls per year were “Adverse reaction” cases,  primarily to “Food” and “Drugs,” but also to “Other” which includes dietary supplements.)

As defined in Part I, the GAO report[3] did not include dietary supplements intended for veterinary usage or those that were available only by prescription (e.g., prescription only folic acid and fish oil supplements). The GAO did, however, include products sold as dietary supplements that were deliberately adulterated (i.e., with active pharmaceutical ingredients and/or their analogs or banned substances such as ephedra). Although these products were included by the GAO because they were marketed as dietary supplements to consumers who were not aware they were not legitimate supplements or could cause serious health problems, it is for this very reason the products should be excluded. In our opinion, the inclusion of intentionally adulterated products (which we included in our Part I count to accommodate the GAO’s position) equates to illegal actions being charged to the accounts of legitimate companies.[7] For this reason, deliberately adulterated products will be excluded from further consideration in this article. In addition, homeopathics and cultural medicines[8] which are not considered by FDA to be dietary supplements have also been excluded, as has caffeine when used as an over-the-counter type stimulant product, but not when ingested as an ingredient in energy drinks or multi-ingredient dietary supplements.

The GAO included only PC cases in which an individual had an adverse reaction when the dietary supplement was used as directed. (No intentional misuse or abuse, or unintentional incidents were included in the GAO report because FDA had stated that SAERs generally covered products that were being used as directed). To tease out this information, the GAO accessed files that are not openly available to the public or part of the annual reports. During our examination, we have had access only to the annual reports. Based on the revised categorization of dietary supplements, the NPDS report data for 2008 – 2011 now yield a total of 13,222 “Adverse reaction” exposure cases associated with dietary supplements, an annual average of 3306 (3133 in 2008, 3203 in 2009, 3327 in 2010, 3559 in 2011). In Figure 1 the relationship of dietary supplement “Adverse reaction” calls (for products taken according to label directions) to “Intentional” and “Unintentional” exposure calls and to “All dietary supplement exposures” is presented for calls logged by the AAPCC from 2008 – 2011[9]. These calls include all case outcomes, from those having no outcome or effect to death. Each of the following dietary supplement product categories had at least 100 calls logged as “Adverse reaction” in a single year:

8619 Vitamin B3 (Niacin)

297 Multi-botanicals without ma huang or citrus aurantium

184 Iron and iron salts (excluding vitamins with iron)

171 Other single ingredient botanicals

127 Adult formulations vitamins without iron or fluoride

127 Vitamin D

108 Multi-mineral and multi-herbal dietary supplements

98 Adult formulation vitamins with iron (but no fluoride)

86 Other B complex vitamins (i.e., not B3 niacin or B6 pyridoxine alone)

81 Energy drinks (caffeine only; without guarana, kola nut, tea, yerba mate, cocoa, etc.)

54 Miscellaneous (other types of) vitamins

Figure 1. Logged dietary supplement calls by year. Dietary supplement calls have been revised to exclude homeopathies, cultural medicines, OTC medicines, and illicit or adulterated products. Calls for all outcomes (none, minor, moderate, major, and death) are included.

As explained in Part I, although logged AAPCC calls for confirmed human exposures to dietary supplements are sifted to separately report cases addressing misuse, abuse, or accidental ingestion from cases associated with adverse reaction and although the same calls are also sorted regarding outcome (which can range from no outcome to death), the AAPCC reporting system provides no generally available means of linking the exposure type to outcome. Only in cases in which death is an outcome is the type of exposure specified (in the case reports). The outcomes for all categorizable dietary supplement exposures (2008-2011) are presented pictorially in Figure 2[10]. The number of outcomes for the four-year period totals 107,208. Exposures resulting in no effect or only a minor effect account for 95.95% (102,865) of the outcomes. Exposures resulting in moderate effects (i.e., signs or symptoms that often require some form of treatment, but are not life-threatening, disabling or disfiguring) account for 3.85% (4127) of the outcomes. Exposures resulting in a major effect (i.e., signs or symptoms that are life-threatening, disabling or disfiguring) or death (as a direct result or complication) account for 0.19% (201) and 0.01% (15) of the outcomes, respectively. As a simple proportion based on occurrence, approximately 3% of each outcome category could be attributed to adverse reactions to dietary supplement products. When one considers that “Intentional” exposures include the subcategories of misuse, abuse, and suicide and “Unintentional” exposures include therapeutic errors or incorrect administrations all of which carry a greater likelihood of a severe outcome, the 3% approximation is seen not to be a conservative estimate. In descending order, the dietary supplement categories that were associated with ≥ 50 moderate outcomes in a single year were:

1569 Multi-botanicals without ma huang or citrus aurantium

116 Vitamin B3 (Niacin)

103 Iron and iron salts (excluding vitamins with iron)

71 Multi-mineral and multi-herbal dietary supplements

43 Sodium and sodium salts

42 Other single ingredient botanicals

39 Energy drinks: Caffeine only (without guarana, kola nut, tea, yerba mate, cocoa, etc.)

33 Energy drinks: Caffeine containing (from any source including guarana, kola nut, tea, yerba mate, cocoa, etc.)

In descending order, the categories associated with at least five major outcomes in a single year were:

69 Multi-botanicals without ma huang or citrus aurantium[11]

6 Iron and iron salts (excluding vitamins with iron)

4 Vitamin B3 (Niacin)

3 Other single ingredient botanicals

In the AAPCC NPD annual reports, all fatal outcomes are tabulated with case reports filed on every death for which a documented exposure is determined to be a contributing factor.[12] A brief summary of those cases in which a dietary supplement product was found present (contributing or not) follows:

  1. In 2008, dietary supplements were identified as present[13] in 12 out of 1315 human fatalities. In ten cases, causality was dismissed, i.e., “No relative contribution to fatality.” The other two cases were “Intentional” exposures (suspected suicide). In one, caffeine (weight loss product) was identified as being causative (“Undoubtedly responsible”). In the other, guarana was identified along with cocaine and the mixture was viewed as “Undoubtedly responsible.” None of these cases qualify as “Adverse reaction.”

  2. In 2009, dietary supplements were present in 17 out of 1158 mortality cases. In 14 cases, causality was dismissed. In one case, an “Intentional” exposure (suspected suicide), caffeine was identified as being one component in in an eight component mixture that included street drugs; the mixture was determined to be “Undoubtedly responsible.” In two additional cases, dietary supplements were found to be “Probably responsible” (iron) or “Contributory” (senna, an herb); these cases were both “Intentional” exposures. None of these cases met the definition of “Adverse reaction.”

  3. In 2010, dietary supplements were identified in 18 fatal cases (1146 total). Causality was dismissed in 12 cases. In three additional cases, dietary supplement products (iron, magnesium sulfate, potassium chloride) were identified as being “Undoubtedly responsible” but all three cases were “Intentional” suspected suicides. In the three other cases, two dietary supplement type materials and a dietary supplement were found to be “Probably responsible” (selenium) and “Contributory” (sodium chloride and an unspecified dietary supplement). The selenium case was an “Unintentional” occupational exposure that may have involved inhalation of Lselenomethione powder. The sodium chloride exposure involved a five-year old child and was categorized as “Unknown” due to a lack of understanding regarding the circumstances. The unspecified dietary supplement exposure was determined to be “Contributory” in an “Adverse reaction” and so counts in our evaluation.

  4. In 2011, dietary supplements were identified as present in 23 out of 1995 human fatalities. Causality was dismissed in 19 cases. In one of the remaining cases, an “Adverse reaction” exposure, a dietary supplement product (magnesium sulfate) was identified as being “Undoubtedly responsible.” In three additional cases, dietary supplement products were found to be “Probably responsible” (sodium bicarbonate) or “Contributory” (caffeine energy drink and Vitamin B12). Exposure to sodium bicarbonate was due to an “Unintentional” therapeutic error and acute exposure to Vitamin B12 was due to “Intentional” misuse. Although the caffeine energy drink was found to contribute to a death in an “Adverse reaction” exposure to food (AAPPC categorization), we will include this incident in our evaluation as an “Adverse reaction” to a dietary supplement product. In 2011, two fatalities were due to “Adverse reaction” exposures to dietary supplement products, one being “Undoubtedly responsible” and one being “Contributory.”

From 2008 – 2011, 13,222 “Adverse reaction” exposure calls were logged by the AAPPC regarding dietary supplement products, but dietary supplements , when used as directed, were “Undoubtedly responsible” for only one human fatality (magnesium sulfate) and “Contributory” in only to two additional deaths (unspecified dietary supplement and caffeine energy drink). In summary, SAEs do not appear to be under-reported by the dietary supplement industry; instead this evaluation demonstrates that true dietary supplements (i.e., products that have not been deliberately adulterated) taken in accordance with label instructions are amazingly safe.

Figure 2. Outcomes for all dietary supplement exposures (2008-2011) including intentional, unintentional, other, and adverse reaction). Approximately 3% of each outcome could be associated with an adverse reaction exposure.



Bronstein, A.C., Spyker, D.A., Cantilena, Jr., L.R., Green, J.L., Rumack, B.H., and Giffen, S.L. (2009) 2008 Annual report of the American Association of Poison Control Centers’ National Poison Data System (NPDS): 26th Annual report. Clinical Toxicology, 47: 911-1084.

Bronstein, A.C., Spyker, D.A., Cantilena, Jr., L.R., Green, J.L., Rumack, B.H., and Giffen, S.L. (2010) 2009 Annual report of the American Association of Poison Control Centers’ National Poison Data System (NPDS): 27th Annual report. Clinical Toxicology, 48: 979-1178.

Bronstein, A.C., Spyker, D.A., Cantilena, Jr., L.R., Green, J.L., Rumack, B.H., and Dart, R.C. (2011) 2010 Annual report of the American Association of Poison Control Centers’ National Poison Data System (NPDS): 28th Annual report. Clinical Toxicology, 49: 910-941 with on-line appendix materials (tabulated data) pp. 1-171.

Bronstein, A.C., Spyker, D.A., Cantilena, Jr., L.R., Green, J.L., Rumack, B.H., and Dart, R.C. (2012) 2011 Annual report of the American Association of Poison Control Centers’ National Poison Data System (NPDS): 29th Annual report. Clinical Toxicology, 50: 911-1164


[1] For an AAPCC call to be logged as an ‘Adverse reaction’, the product was being used as directed (which excludes ‘Unintentional’ exposures and ‘Intentional’ misuses).

[2] Common causes of SAEs found not to be related to dietary supplements include pre-existing health condition(s) and/or other products ingested at the same time.

[3] GAO, “Dietary Supplements: FDA may have opportunities to expand its use of reported health problems to oversee products”, GAO-13-244 (Washington, D.C.: March 18, 2013).

[4] No means for assessing potential overlap was available.

[5] Because multiple outcomes may be reported in each SAER (e.g., hospitalization and serious illness), totaling outcome numbers is not meaningful.

[6] Information required in an SAER: (1) an identifiable patient; (2) an identifiable initial reporter; (3) identity and contact information for the responsible person (i.e., the manufacturer, packer, or distributor submitting the serious adverse event report to FDA); (4) a suspect dietary supplement; and (5) a serious adverse event or fatal outcome (“Guidance for Industry: Questions and Answers Regarding Adverse Event Reporting and Recordkeeping for Dietary Supplements as Required by the Dietary Supplement and Nonprescription Drug Consumer Protection Act”, Revised June 2009).

[7] The GAO report noted FDA’s observation that companies that intentionally adulterate supplements with active or illicit ingredients are also noncompliant in meeting AER requirements.

[8] Asian, Ayurvedic, Hispanic or other traditional medicines.

[9] Average number of calls per year from 2008 – 2011.

[10] Outcomes due to ‘Unknown’ adverse reaction exposures are not included because they cannot be categorized.

[11] The only category associated with up to ten major outcomes in a single year (2008).

[12] Because individuals are generally exposed to more than one substance (as food, vitamins, medications), most exposures are found not to be contributing.

[13] Samples in human fatality cases are assayed for a variety of substances including vitamins, etc.

Follow Us

  • LinkedIn
  • Facebook
  • Youtube
  • X

407-802-1400 ext 170

Have a Question?

Keep Up With New Information Released by FDA & EPA          
  • LinkedIn
  • Facebook
  • Twitter
  • Instagram
  • YouTube
bottom of page