Photo of Bexis

We’ve covered a lot of topics since we started DDLaw in late 2006.  One that we haven’t addressed in any real fashion is overwarning – except for an unusual allegation made by a plaintiff that a defendant was “hiding” a warning among too many other warnings.  Cf. Kendall v. Hoffmann-La Roche, Inc., 2010 WL 3034453, at *22 (N.J. Super. App. Div. Aug. 5, 2010) (plaintiff complained that a drug’s warnings focused “predominantly” on birth defects and suicide but only “alluded to” about her condition (“abdominal and bowel problems”) in a “less conspicuous or pointed manner”).

We remedy that today.

Usually, the concept of overwarning is asserted by a defendant – that the plaintiff is threatening the usefulness of warnings about serious, relatively frequent risks by demanding that a defendant include a plethora of warnings about less serious and/or less frequent risks, or otherwise seeks to clutter up the defendant’s warnings with details of one sort or another.  As the Supreme Court observed, albeit not in the product liability context, “[m]eaningful disclosure does not mean more disclosure.  Rather, it describes a balance between competing considerations of complete disclosure and the need to avoid informational overload.”  Ford Motor Credit Co. v. Milhollin, 444 U.S. 555, 568 (1980) (citation and quotation marks omitted).  A typical critique of the problem of overwarning in the general (non-drug/device) product liability context occurred in Liriano v Hobart Corp., in consideration of the “open and obvious” limitation on the duty to warn:

[W]hen a warning would have added nothing to the user’s appreciation of the danger, no duty to warn exists as no benefit would be gained by requiring a warning. . . .  This is particularly important because requiring a manufacturer to warn against obvious dangers could greatly increase the number of warnings accompanying certain products. . . .  Requiring too many warnings trivializes and undermines the entire purpose of the rule, drowning out cautions against latent dangers of which a user might not otherwise be aware.  Such a requirement would neutralize the effectiveness of warnings as an inexpensive way to allow consumers to adjust their behavior based on knowledge of a product’s inherent dangers.

700 N.E.2d 303, 308 (N.Y. 1998) (citation omitted) (emphasis added).  Here’s another:

[T]he proliferation of label detail threatens to undermine the effectiveness of warnings altogether.  As manufacturers append line after line onto product labels in the quest for the best possible warning, it is easy to lose sight of the label’s communicative value as a whole.  Well-meaning attempts to warn of every possible accident lead over time to voluminous yet impenetrable labels-too prolix to read and too technical to understand.

Hood v. Ryobi America Corp., 181 F.3d 608, 611 (4th Cir. 1999) (applying Maryland law) (emphasis added).  See Johnson v. American Standard, Inc., 179 P.3d 905, 914 (Cal. 2008) (“[r]equiring manufacturers to warn their products’ users in all instances would place an onerous burden on them and would invite mass consumer disregard and ultimate contempt for the warning process”); Thompson v. County of Alameda, 614 P.2d 728, 735 (Cal. 1980) (excessive warnings “produce a cacophony . . . that  by reason of their sheer volume would add little to the effective protection of the public”); Cotton v. Buckeye Gas Products Co., 840 F.2d 935, 937-38 (D.C. Cir. 1988) (the “primary cost [of overwarning] is . . . the increase in time and effort required for the user to grasp the message.  The inclusion of each extra item dilutes the punch of every other item. . . .  [W]e must review the record in light of these obvious information costs”); Aetna Casualty & Surety Co. v. Ralph Wilson Plastics Co., 509 N.W.2d 520, 523 (Mich. App. 1993) (“excessive warnings on product labels may be counterproductive, causing ‘sensory overload’ that literally drowns crucial information in a sea of mind-numbing detail”); Broussard v. Continental Oil Co., 433 So. 2d 354, 358 (La. App. 1983) (“As a practical matter, the effect of putting at least ten warnings on the [product] would decrease the effectiveness of all of the warnings. A consumer would have a tendency to read none of the warnings if the [product] became cluttered with the warnings”); Heckman v. Ryder Truck Rental, Inc., ___ F. Supp.2d ___, 2013 WL 4478690, at *8 (D. Md. Aug. 21, 2013) (“[t]he price of more detailed warnings is greater than their printing fees; it also takes into account whether additional details will undermine the effectiveness of warnings altogether”); Stanley v. Central Garden & Pet Corp., 891 F. Supp. 2d 757 (D. Md. 2012) (same); New York State Pesticide Coalition, Inc. v. Jorling, 704 F. Supp. 26, 29-30 (N.D.N.Y. 1989) (“[p]lacing every potential warning or use requirement onto the label could operate to dilute the most important instructions”), aff’d, 874 F.2d 115 (2d Cir. 1989).

Overwarning has also been the subject of a fair amount of scholarly treatment.  The Third Restatement of Torts recognizes that excessive numbers of warnings “may be ignored by users and consumers and may diminish the significance of warnings about [other] risks” and “could reduce the efficacy of warnings generally.”  Restatement (Third) of Torts, Products Liability §2, comment j (1998).  Prosser also addresses the problem of overwarning, considering it “a naive belief that one can warn against all significant risks.  Too much detail can be counterproductive.”  W. Page Keeton, et al., Prosser & Keeton on the Law of Torts §96, at 686 (5th ed. 1984).  Commentators have made the same point.

Whether the additional warnings take some attention away from the warnings already given, establish the misconception that all risks are the same, or lack content which would allow the user to avoid the risk, are all issues that must be considered as part of the cost inherent in providing additional warnings.  From the standpoint of overall safety, providing an additional warning does not by itself serve the end of risk minimization if its effect is to denigrate the standing of other warnings which are more important in the safe utilization of the product.

Theodore S. Jankowski, “Focusing on Quality & Risk:  The Central Role of Reasonable Alternatives in Evaluating Design & Warnings Decisions,” 36 S. Tex. L. Rev. 283, 298 (1995).

Overwarning effectively amounts to the manufacturer “crying wolf.”  The more that product manufacturers warn of risks that never materialize, the less likely product users are to heed those warnings.  If product users merely ignored the excessive warnings, the problem might be minimal − the only superfluous costs would be those of providing the warnings.  However, product users might begin to ignore not only the excessive warnings, but also those that are crucial to safe product use.  As warning labels become increasingly crowded with seemingly redundant, useless admonitions to avoid subjecting oneself to obvious risks, product users and consumers may increasingly view all warnings as similarly useless.

Robert G. Knaier, “An Informed-Choice Duty to Instruct?” 88 Cornell L. Rev. 814, 853 (2003).  See also Jeff Todd, “A Rhetoric of Warning Defects,” 54 S. Tex. L. Rev. 343, 374, 377 (2012) (“inclusion of the unnecessary and the inane can reduce the efficacy of warnings by diminishing the significance of warnings about non-obvious, not-generally-known risks”; “the aggregation of nonmaterial disclosures will make the cost of reading and remembering outweigh the benefits to the average user, who will not read them”); James A. Henderson, Jr. & Aaron D. Twerski, “The Products Liability Restatement in the Courts:  An Initial Assessment,” 27 Wm. Mitchell L. Rev. 7, 16, 19 (2000) (requiring “warnings about obvious and well known risks diminish the significance of warnings and tend to clutter warning labels with useless information”); Frederick C. Schafrick, “Product Liability Suits for Failure to Warn of the Hazards of Regulated Products,” 32 Tort & Ins. L.J. 833, 858 (1997) (“there are risks from overwarning.  When overwarning results in products with smaller risks containing warnings comparable to those on products with greater risks, the information provided consumers has been made less useful”); W. Kip Viscusi, “Individual Rationality, Hazard Warnings, & the Foundations of Tort Law, 48 Rutgers L. Rev. 625, 665-66 (1996) (“[e]xcessive warnings are not innocuous”; “If warnings indicate a high relative risk when there is none, they will distort relative product comparisons, thus compromising credibility” because they “further dilute the warnings for the real hazards.”); Michael S. Jacobs, “Toward a Process-Based Approach to Failure-to-Warn Law,” 71 N.C. L. Rev. 121, 158 (1992) (overwarning subjects users to “the perplexing and unpleasant prospect of a regime of warnings that are sometimes incomplete and sometimes too complete”); William H. Hardie, Jr., “Foreseeability:  A Murky Crystal Ball for Predicting Liability,” 23 Cumb. L. Rev. 349, 408 (1993) (“[t]he meaning of ‘foreseeable’ is so uncertain that overwarning will surely result if manufacturers are required to warn of hazards attributable to merely foreseeable uses”); James A. Henderson, Jr., “Doctrinal Collapse in Products Liability: The Empty Shell of Failure to Warn, 65 N.Y.U. L. Rev. 265, 327 n.135 (1990) (“Overwarning causes users and consumers to discount or ignore warnings that should be heeded, leading to higher accident costs which, though very real, are not before the court in failure-to-warn litigation.  Overwarning also may scare some worthwhile users away, resulting in wastefully high avoidance costs.”); Michael D. Green, “When Toxic Worlds Collide:  Regulatory & Common Law Prescriptions for Risk Communication,” 13 Harv. Envtl. L. Rev. 209, 223 (1989) (“[m]uch like the little boy who cried wolf, myriad warnings that surround everyone and often call attention to trivial or well-known risks tend to reduce the attention that is paid to all warnings, thereby reducing their overall effectiveness”); Victor E. Schwartz, “Warnings in the Workplace:  The Need for A Synthesis of Law & Communication Theory,” 52 U. Cin. L. Rev. 38, 54, 60 (1983) (“the “product user may be confused or overwhelmed . . . or unable to assimilate effectively the necessary information about the most important hazards”; “[i]f every possible danger in life were accompanied by a warning, product users quickly would become inured to all warnings, and eventually would ignore them”); Aaron D. Twerski, et al., “The Use & Abuse of Warnings in Products Liability − Design Defect Litigation Comes of Age, 61 Cornell L. Rev. 495, 513-17 (1976) (“warnings, in order to be effective, must be selective. . . .  If even remote risks are to be forced to the consumer’s attention, the danger signal is diluted”).  Interestingly, though, the earliest legal reference we’ve found to overwarning was from Congress, when it was considering hazardous substance labeling over fifty years ago:

If labeling were required to caution against the risk of even the most trifling indisposition, there would hardly be any substance . . . which would not have to bear cautionary labeling, so that consumers would tend more and more to disregard label warnings, thus inviting indifference to cautionary statements on packages of substances presenting a real hazard of substantial injury or illness.

H.R. Rep. No. 86-1861 (1960), reprinted in, 1960 U.S.C.C.A.N. 2833, 2837.  We don’t usually think of Congress as prescient (what’s the opposite of progress?), but in this instance it was.

Closer to home, similar concerns have been expressed that allowing plaintiff to demanding warnings and instructions about this, that, and the other thing also runs the risk of reducing the effect of warnings about drugs and medical devices.  The California Supreme Court has referred to risks of overwarning in the prescription medical product context several times.  In Finn v. G.D. Searle & Co., 677 P.2d 1147 (Cal. 1984), the court held that it was improper to instruct a jury that every drug risk should be warned about:

[I]t seems obvious that liability ought not to be imposed for failure to warn based on every piece of information in a manufacturer’s possession.  The quality of evidence supporting a causal connection between product and injury may range from extremely vague to highly certain. . . .  If we overuse warnings, we invite mass consumer disregard and ultimate contempt for the warning process.  Moreover, both common sense and experience suggest that if every report of a possible risk, no matter how speculative, conjectural, or tentative, imposed an affirmative duty to give some warning, a manufacturer would be required to inundate physicians indiscriminately with notice of any and every hint of danger, thereby inevitably diluting the force of any specific warning given.

Id. at 1153 (citations and quotation marks omitted).  Next, in Ramirez v. Plough, Inc., 863 P.2d 167 (Cal. 1993), the court similarly held that overwarning concerns weighed against requiring that that drug warnings to be given in languages other than English:

The burden of including warnings in so many different languages would be onerous . . . and at some point might prove ineffective or even counterproductive if the warning inserts became so large and cumbersome that a user could not easily find the warning in his or her own language.

Ramirez v. Plough, Inc., 863 P.2d 167, 175-76 (Cal. 1993).  Most recently, in Dowhal v. SmithKline Beecham Consumer Healthcare, 88 P.3d 1 (Cal. 2004), a medical device case, the court reiterated, “a truthful warning of an uncertain or remote danger may mislead the consumer into misjudging the dangers stemming from use of the product, and consequently making a medically unwise decision.”  Id. at 14.

The same considerations that repeatedly motivated the California Supreme Court in prescription medical product cases have also been articulated by other appellate courts.  “If pharmaceutical companies were required to warn of every suspected risk that could possibly attend the use of a drug, the consuming public would be so barraged with warnings that it would undermine the effectiveness of these warnings.”  Doe v. Miles Laboratories, Inc., 927 F.2d 187, 194 (4th Cir. 1991) (applying Maryland law).  See Doe v. Miles, Inc., 2000 WL 667383, at *16 (Mo. App. May 23, 2000) (quoting and following Fourth Circuit Doe v. Miles decision).  The earliest appellate rejection of a pharmaceutical warning claim on the basis of overwarning concerns was back in 1983 by a Michigan court.  Its reasoning is equally apt today:

[A] “danger” is many things to different people.  Especially with drugs, the danger can vary from person to person.  Some people are more sensitive to certain drugs than are others; thus, methods of avoiding the danger may vary depending on the user’s susceptibility to danger and the type of reaction suffered.  This leads to another reason:  sensory overload.  Commentators have cautioned that excessive warnings on product labels may be counterproductive.

Dunn v. Lederle Laboratories, 328 N.W.2d 576, 580-581 (Mich. App. 1983) (citing the 1976 Twerski article mentioned above).

Considerably more recently, Judge Posner rejected an argument that the defendant should have warned about “rashes” as precursors to serious skin reactions:

But then the label would have had to describe as well every other serious disease that might, however infrequently, be caused, or even just arguably caused (for it is unclear whether [this drug] can cause [the condition]), by [the product].  And it would have to recite the symptoms of the disease if it was rare.  The resulting information overload would make label warnings worthless to consumers.

Robinson v. McNeil Consumer Healthcare, 615 F.3d 861, 869 (7th Cir. 2010) (applying Virginia law).

Warnings about dangers with less basis in science or fewer hazards could take attention away from those that present confirmed, higher risks.  A label with many varied warnings may not deliver the desired information to users.  Space on product labeling material is also a factor, and the most effective labels are those with large, bold warnings and a simple design.

Brooks v. Howmedica, Inc., 273 F.3d 785, 796-797 (8th Cir. 2001) (en banc) (applying Minnesota law).  Likewise in Thomas v. Hoffmann-LaRoche, 949 F.2d 806 (5th Cir. 1992) (applying Mississippi law), the court noted:

[P]erhaps worse, we would force drug manufacturers to list, and perhaps contraindicate, every possible risk in order to avoid the possibility of liability.  If manufacturers so respond to the possibility of liability, physicians will begin to ignore or discount the warnings provided by the drug manufacturers.  Permitting a jury to find liability on such a basis would undermine the important role of warnings as a device to communicate vital information to physicians.

Id. at 816 n.40.

A raft of trial courts agree.  Georges v. Novartis Pharmaceuticals Corp., 2013 WL 5217198, at *5 (C.D. Cal. April 4, 2013) (“evidence relating to overwarning in general is admissible to demonstrate that Defendant considered overwarning when formulating [its] labels”); Mahaney v. Novartis Pharmaceuticals Corp., 835 F. Supp.2d 299, 321 (W.D. Ky. 2011) (“[o]verwarning, or warning fatigue, is a legitimate concern of manufacturers when creating labels for consumers”); Hogan v. Novartis Pharmaceuticals Corp., 2011 WL 1533467, at *14 (E.D.N.Y. April 23, 2011) (“mentioning too many warnings of serious injuries would have diluted the effectiveness of warnings generally . . . goes to the heart of the case”) (applying Rhode Island law); Gaeta v. Perrigo Pharmaceuticals Co., 562 F. Supp.2d 1091, 1097 (N.D. Cal. 2008) (“[o]verwarning, just like underwarning, can similarly have a negative effect on patient safety and public health”; “attempts to impose additional warnings can lead to labeling that does not accurately portray a product’s risks”) (quoting FDA), rev’d on other grounds, 630 F.3d 1225 (9th Cir. 2011) (no generic drug preemption), vacated, 132 S.Ct. 497 (2011) (generic drug preemption); Sykes v. Glaxo-SmithKline, 484 F. Supp.2d 289, 309 (E.D. Pa. 2007) (“[t]o allow the plaintiffs’ warnings would lead to the very result the FDA wants to avoid, i.e., overwarning, exaggeration, and defensive labeling”); Ames v. Apothecon Inc., 431 F. Supp.2d 566, 573 (D. Md. 2006) (“warnings must be brief and focused to be effective”); Orso v. Bayer Corp., 2006 WL 2794975, at *3 (N.D. Ill. Sept. 27, 2006) (“the duty that plaintiff advocates seems likely to inundate physicians with information they do not need and cannot use”); In re Bextra and Celebrex Marketing Sales Practices & Products Liability Litigation, 2006 WL 2374742, at *5 (N.D. Cal. Aug. 16, 2006) (“additional requirements for the disclosures or risk information are not necessarily more protective of patients. Instead, they can erode and disrupt the careful and truthful representation of benefits and risks that prescribers need to make appropriate judgments about drug use. Exaggeration of risk could discourage appropriate use of a beneficial drug”) (quoting FDA); Andrulonis v. United States, 724 F. Supp. 1421, 1500 (N.D.N.Y. 1989) (“drafters of a package insert must be mindful that placing every potential warning or use requirement onto the label could operate to dilute the most important instructions”) (citation and quotation marks omitted), aff’d in pertinent part, rev’d in part on other grounds, 924 F.2d 1210 (2d Cir. 1991), vacated mem., 502 U.S. 801 (1991); Edmondson v. International Playtex, Inc., 678 F. Supp. 1571, 1574-75 (N.D. Ga. 1987) (“A warning, whether by an insert or directly on the package, is most effective if it is concise and reasonably brief inasmuch as the ultimate goal is to have it read by users.  The grant to judges and juries of the right to compel additional language is not conducive to brevity.”); Gourdine v. Crews, 2006 WL 5277412 (Md. Cir. June 28, 2006) (“[o]verstating risks can discourage appropriate use of a drug, and can cause risks to lose their significance, so that none are seriously considered”), aff’d, 935 A.2d 1146 (Md. App. 2007), aff’d on other grounds, 955 A.2d 769 (Md. 2008).

As many of the above cases recognize, the overwarning problem also has a regulatory component. For over 30 years, the FDA itself has had plenty to say about the issue.   Back in 1975 (when Gerald Ford was president, and Bexis was just starting college), the FDA cautioned that unsubstantiated risk information in labeling “would result in such uncertainty and confusion that the usefulness of [existing] warnings in protecting the public against possible harm would be severely undermined, if not destroyed.” 40 Fed. Reg. 28582, 28583 (FDA July 7, 1975).

Physicians are always in a position to pursue additional information through normal educational sources, such as treatises and medical journals. . . .  [T]he Commissioner does not agree that general statements on good professional practice are appropriate for drug labeling.  There are potentially many such statements, which, if all were included in drug labeling, would transform labeling into small text-books of medicine.  As a general policy, therefore, these regulations will not require such statements to be included in labeling.

44 Fed. Reg. 37434, 37436 (FDA June 26, 1979).

The 21st Century FDA agrees.  Its current guidance on how to prepare effective medical device warnings states:

Including too many warnings and precautions, over-warning, dilutes the strength of all of the hazard alerts.  We recommend that writers use care in what is designated as a warning or precaution.  Careless designation can have the same diluting effect as over-warning . . . .  Repeated exposure to unnecessary hazard alerts (not relevant or already known) reduces the effectiveness of the important warnings and precautions.

FDA CDRH, Guidance on Medical Device Patient Labeling: Final Guidance for Industry and FDA Reviewers, at 42 (2001) (available here).   See also 73 Fed. Reg. 2848, 2851 (FDA Jan. 16, 2008) (“[e]xaggeration of risk, or inclusion of speculative or hypothetical risks, could discourage appropriate use of a beneficial drug, biologic, or medical device”); 71 Fed. Reg. 3922, 3935, 3927 (FDA Jan. 24, 2006) (“labeling that includes theoretical hazards not well-grounded in scientific evidence can cause meaningful risk information to lose its significance”; “[o]verwarning, just like underwarning, can similarly have a negative effect on patient safety and public health”; “FDA believes that including relative or hypothetical hazards [as contraindications] diminishes the usefulness of this section.”); FDA, CDER,“Draft Guidance for Industry: Brief Summary:  Disclosing Risk Information in Consumer-Directed Print Advertisements,” at 2 (Jan. 2004) (“exhaustive lists of minor risks distract from and make it difficult to comprehend and retain information on the more important risks”) (available here); FDA CDRH, “Write it Right: Recommendations for Developing User Instruction Manuals for Medical Devices Used in Home Health Care,” at 7 (Aug. 1993) (“Overwarning has the effect of not warning at all. The reader stops paying attention to excess warnings.”); 53 Fed. Reg. 30522, 30530 (FDA Aug. 12, 1988) (“The agency agrees that too many warning statements reduce the impact of important statements.”); 44 Fed. Reg. at 37447 (risks should be included in labeling “only when reasonable evidence exists indicating an association between the drug and a hazard. . . . [I]ncluding theoretical hazards as contraindications in drug labeling would cause that very important section of the labeling to lose its significance”); 43 Fed. Reg. 1101, 1104 (FDA Jan. 6, 1978) (FDA “does not believe that a warning or cautionary statement should be required for every possible question that might be raised about the safety of a product. A plethora of warnings about insubstantial questions would be difficult for consumers to evaluate.”).

These views are shared by commentators on FDA regulated products.  A recent survey determined that an average prescription drug package insert lists 49 potential adverse drug events, and ten percent of these list over 500.  Jon Duke, Jeff Friedlin, Patrick Ryan, “A Quantitative Analysis of Adverse Events and ‘Overwarning’ in Drug Labeling,” 171 Arch. of Int. Med. 944, 945 (2011). Excessive warnings are not helpful to physicians:

[O]verwarning of prescription drug side effects may adversely affect treatment decisions if other options (which fall outside of the products liability system) do not have to carry equally alarming risk information.  The FDA has an interest in rational prescribing . . . so that a physician can compare them with other available therapies.  That goal is not advanced if a drug is made to appear riskier than other drugs and other therapies due to the over-dramatization of risk information.  To allow a warning based on inconclusive evidence or scientific hunches results in doctors not prescribing effective drugs to a patient because of the erroneous belief that a side-effect might occur.  Alternatively, physicians may tune out if overwhelmed with risk information. In either case, the result may be suboptimal treatment choices.

Lars Noah, “Medicine’s Epistemology:   Mapping The Haphazard Diffusion Of Knowledge In The Biomedical Community,” 44 Ariz, L. Rev. 373, 455-56 (2002) (footnotes, citations, and quotation marks omitted); accord Lars Noah, “The Imperative to Warn:  Disentangling the ‘Right to Know’ from the ‘Need to Know’ About Consumer Product Hazards,” 11 Yale J. Reg. 293, 382-83 (1994) (“[i]n the event that labeling included warnings of all possible side effects, the cacophony of risk information could undermine a doctor’s ability to appreciate warnings about meaningful hazards”); W. Kip Viscusi, et. al., “Deterring Inefficient Pharmaceutical Litigation: An Economic Rationale for the FDA Regulatory Compliance Defense,” 24 Seton Hall L. Rev. 1437, 1468 (1994) (“[o]verwarning is also not in the interests of consumers because it hinders the ability of health care professionals to distinguish the relative risks posed by various drugs”); Thomas Scarlett, “The Relationship Among Adverse Drug Reaction Reporting, Drug Labeling, Product Liability, and Federal Preemption,” 46 Food & Drug L.J. 31, 36 (1991) (“overstated warnings could tip the judgment of the medical profession in an undesirable direction”); Richard A. Epstein, “Legal Liability for Medical Innovation,” 81 Cardozo L. Rev. 1139, 1150 (1987) (“[t]he full costs of overwarning would only be known if legal actions were available to people deterred from taking needed therapy by excessive warnings”; ); Richard M. Cooper, Drug Labeling & Products Liability: The Role of the Food and Drug Administration, 41 Food & Drug L.J. 233, 238 (1986) (the goal of rational physician prescribing “is not advanced if a drug is made to appear riskier than other drugs and other therapies due to the over-dramatization of risk information or the presentation of risk information that should not rationally influence prescribing (or treatment) decisions”).

Our failure to devote much attention to overwarning arguments earlier may mean we’re lazy, but it doesn’t mean for a moment that the concerns aren’t real or that the arguments aren’t valid.  As the precedent we’ve discussed demonstrates, in a proper case – particularly where a plaintiff resorts to nitpicking warnings about information that is trivial or tangential – concerns about overwarning can be dispositive.