Enhancing transparency has been one of the FDA’s top priorities for nearly a decade. In late 2016, the agency continued to advance this goal by publishing data relating to food, cosmetics and dietary supplements from its Adverse Event Reporting System, maintained by the Center for Food Safety and Applied Nutrition (CAERS). The CAERS database includes reports of complaints — called “adverse events” — submitted to the FDA. While the FDA has long made adverse event information for drugs accessible to the public on their internet site, until recently, adverse event reports for food, cosmetics and dietary supplements in the CAERS database were not published and were only accessible to the public through a FOIA request.
This article discusses the various FDA adverse event reporting systems and the limitations of adverse event reports in determining causality, and also explores the ways in which adverse event reports have been used in pharmaceutical litigation. This litigation foreshadows how these reports may be used against food, supplement and cosmetics manufacturers in the future.
I. The FDA’s Adverse Event Reporting Systems
The CAERS database captures reports of adverse events submitted to the FDA that relate to food, cosmetics, and dietary supplements. The term “adverse events” refers to complaints of both minor and major medical events, as well as complaints concerning defective packaging, labeling or other non-medical issues. FDA regulations define an adverse event merely as an event associated in proximity of time with the use of the product, “whether or not considered drug related.” These reports may be received directly from healthcare professionals or consumers, or indirectly through the product manufacturer. The published data from CAERS, which can be found on the FDA website, includes basic information about the affected person (sex, age), identification of the product at issue (sometimes with the brand, sometimes without), the dates of the event and the report, and a brief description of the alleged adverse event.
FDA uses CAERS for a variety of purposes, including evaluating a manufacturer’s compliance with reporting requirements, responding to requests for information and identifying or investigating potential safety concerns relating to a product. If the FDA detects a potential safety concern in the CAERS data, the agency may take regulatory action, including issuing a safety alert to the public, or removing a product from the market.
For food and cosmetic manufacturers, adverse event reporting is voluntary. But manufacturers of dietary supplements must report serious adverse events (i.e., those resulting in death, a life-threatening experience, inpatient hospitalization, a persistent disability or incapacity, a congenital anomaly or birth defect, or medical or surgical intervention) that come to their attention.
The FDA maintains a separate Adverse Event Reporting System to collect data on drugs and biological products called FAERS. There are a few key difference between FAERS and CAERS data. First, FAERS safety reports are more difficult to access and require knowledge of relational databases to analyze the data. A simple search for an event or a product cannot be performed. CAERS data are accessible via an Excel spreadsheet that can be downloaded and reanalyzed. Second, manufacturer reporting for drugs and biologics is mandatory. FDA regulations require pharmaceutical and biologic manufacturers to submit adverse event reports whenever they receive information that a patient taking one of their products has experienced an adverse event, whether or not the manufacturer believes it to be causally related to the medication. FAERS also contains reports submitted voluntarily by consumers and physicians through the FDA’s MedWatch program. Like the CAERS data, FAERS reports contain basic demographic information about the injured person, the data of the event and report, and the reporter’s assessment of causation.
II. Limitations of Adverse Event Reports
Adverse event reports, whether relating to drugs, food or cosmetics, have several limitations. First, anyone — including attorneys representing allegedly injured clients — can report an adverse event to the CAERS or FAERS databases, and information contained in the database appears exactly as it was reported to the FDA. Reports come from people with specialized knowledge about medicine or the product — such as healthcare providers or manufacturers, as well as everyday consumers.
Second, the data contain unverified assessments of causation. One of the data fields in the CAERS safety reports asks the reporter’s view of whether the product played a role in the alleged adverse event. Such opinion may be uninformed, speculative or based on incomplete information. This opinion is not typically validated, edited or changed by the FDA. 
Third, the reports do not necessarily include all relevant data, nor do they always include accurate contact information to enable the FDA to follow up. Critical information contained in individual reports may be incomplete or missing, such as underlying medical conditions or the consumer’s use of other potentially relevant products or medications, that may bear on what caused the event at issue.
Fourth, certain events may be overrepresented. Because anyone can report an adverse event (e.g., a consumer and his doctor), CAERS may contain duplicate reports for the same event. The FDA removes duplicate reports it detects. But the FDA’s ability to find duplicate reports depends on the quality and reliability of the data provided, which vary among reports and reporters. Reports can be stimulated based on external factors, causing an event to be overrepresented. For example, publicity concerning the product or a potential health risk can stimulate adverse event reporting. In the pharmaceutical context, newer medications or those with more intrusive routes of administration are more likely to be the subject of reporting. Conversely, adverse events can also be underreported because reporting depends on the reporter — i.e., whether the consumer or medical professional identifies and reports an adverse event. According to the FDA, estimates suggest that the agency only receives reports of roughly 1-10 percent of adverse events that occur.
As acknowledged by the FDA, these limitations hamper the ability to use adverse event reports to make conclusions about causality. Indeed, in releasing the CAERS data, FDA cautioned that there is “no certainty that a suspected product caused a reaction.” But despite these limitations, and the FDA’s explicit warning that adverse event reports cannot solely support a finding of causation, these reports are relied on by plaintiffs’ attorneys and their experts in litigation facing pharmaceutical and device companies, as described in Section III. It is likely that with the release of the CAERS data, manufacturers in the dietary supplement, food and cosmetics industry may face the same challenges.
III. Use of Adverse Event Data in Litigation
With the release of CAERS, information previously available only through a FOIA request will now be readily available to the general public, including researchers, healthcare providers, consumers and lawyers. The availability of this data may stimulate personal injury and consumer litigation. Plaintiffs’ attorneys can use CAERS data — as they have consistently used FAERS data against pharmaceutical manufacturers — to identify adverse events that could form the basis of litigation. One of the ways they can do this is by detecting trends in reporting for specific adverse events. For example, a plaintiff’s attorney (or their expert) could sort the CAERS data by adverse event and look at the most commonly reported adverse event for a given product. Once identified, an adverse event may provide a lure to attract potential clients and class representatives. And the advertising itself may trigger reporting of even more adverse events.
Plaintiffs have also attempted to introduce adverse event reports, and their experts’ analyses of these reports, as evidence of causation. Usually, attempts to do so are unsuccessful. Many courts have held that adverse event reports cannot be used to establish the causation element of a products liability claim. Nevertheless, plaintiffs’ attorneys fight hard to get analyses of adverse event reports — and the underlying reports themselves — admitted into evidence. These documents, they argue, show at the very least that the manufacturer knew or should have identified a safety signal, yet failed to protect patients. Many courts will admit adverse event evidence for this purpose.
But even if admitted for this limited purposes, there is always the risk that jurors may find it hard to grapple with the many limitations inherent in adverse event reporting and so will view the reports and subsequent analyses as solid proof that a product or medication caused the plaintiff’s injury. The manufacturer defendants face an uphill climb to educate jurors on the limitations of adverse event reports to avoid the case being decided against them based on contrived evidence of causation.
Plaintiffs may also use adverse event reports as a tool to ask for additional discovery. This could result in a request for more information on a given event or report, or documents that show how the defendant tracks and analyses reported adverse events. Plaintiffs could also use these reports as a reason to ask for depositions of individuals at the company with knowledge about the reports or the adverse event at issue.
While increased transparency benefits both patients and industry, it also creates risks. The data are available for anyone with a computer to view — including potential plaintiffs and their lawyers. The ready availability of these data presents concerns, especially as adverse events reports for non-pharmaceutical products are on the rise. For example, based on a recent study published in JAMA, adverse events for cosmetic and personal care products are on the rise. Researchers analyzed the CAERS data and found that consumer complaints more than doubled for cosmetic products from 2015 to 2016. Such adverse event data has spurred litigation in the pharmaceutical context. Food, cosmetics and supplement manufacturers should take note of the pharmaceutical industry’s experience and be prepared for similar litigations to come their way. To this end, companies should consider monitoring the data in CAERS to ensure that they are aware of any potential safety or advertising issues that surface before they are served with the first complaint.
About the author
Yvonne M. McKenzie is a partner in the Health Sciences Department of Pepper Hamilton LLP, resident in the Philadelphia office. Ms. McKenzie is a member of the firm’s Food and Beverage Industry Group, is co-leader of the firm's Women's Initiative Network and serves on the Hiring Committee. Ms. McKenzie focuses her practice on counseling and defending companies with products regulated by the FDA, including prescription medications, food, beverages and cosmetics.
 “CAERS” stands for the Center for Food Safety and Applied Nutrition Adverse Event Reporting System.
 21 C.F.R. § 314.80(a) (2009).
 “FAERS” standard for the FDA Adverse Event Reporting System.
 21 C.F.R. § 314.80 (2009).
 See, e.g., In re Zoloft (Sertraline Hydrochloride) Prods. Liab. Litig., 176 F. Supp. 3d 483, 497 (E.D. Pa. 2016); Degidio v. Centocor Ortho Biotecch, Inc., 3 F. Supp. 3d 674, 684 (N.D. Ohio 2014); Drake v. Allegan, Inc., No. 2:13-cv-234, 2014 U.S. Dist. LEXIS 151830 at *21 (D. Vt. Oct. 23, 2014); Caraker v. Sandoz Pharms. Corp., 188 F. Supp. 2d 1026, 1035 (S.D. Ill. 2001); In re Diet Drugs Prods. Liab. Litig., MDL No. 1203, 2001 U.S. Dist. LEXIS 1174, at *48 (E.D. Pa. Feb. 1, 2001);; McClain v. Metabolife Int’l, Inc., 401 F.3d 1233, 1250 (11th Cir. 2005);. But see, Yates v. Ford Motor Co., No. 5:12-cv-752-FL, 2015 U.S. Dist. LEXIS 70476 at * 7-8 (E.D. N.C. May 30, 2015) (refusing to adopt general rule excluding case reports as evidence of causation); Terry v. McNeil-PPC, Inc. (In re Tylenol (Acetaminophen) Mktg., Sales Practices, and Prods. Liab. Litig.), 198 F. Supp. 3d 446, 461 (E.D. Pa. 2016) (finding that case reports alone could not establish causation, but could be use in conjunction with other evidence to support an expert’s causation opinion).
 Michael Kwa, BA, et al., Research Letter, Adverse Events Reported to the US Food and Drug Administration for Cosmetics and Personal Care Products, 177(8):1202-04, JAMA Intern Med. 2017.