We’re always open to guest posts on topics relevant to the defense of prescription medical products. Here’s the first. Sean Wajert is a partner at Jim’s firm, Dechert, and specializes in pharmaceutical defense. It’s about a hot topic – the use and abuse of medical monitoring claims in this type of case.
Introduction
On January 16, 2007, the Appellate Division of the Superior Court of New Jersey handed down a decision in the putative class action case, Sinclair v. Merck, No. A-5661-04T5. The court remanded the matter to the trial court for further proceedings concerning plaintiffs’ claim for the establishment of a court-administered medical screening program, funded by Merck to provide for and/or reimburse the proposed class for certain medical and diagnostic tests. The plaintiffs claimed no present physical injury, but alleged that the cost of diagnostic testing represented a harm or ascertainable economic loss for which they were entitled to compensation.
While I cannot comment directly on the Sinclair case because Dechert is involved in the Vioxx litigation, the decision is but the latest in a series of developments in the controversial area of “medical monitoring.” As this highly unsettled area of the law develops, the ALI in its Council Draft of the Restatement of the Law (Third) Economic Torts and Related Wrongs (Council Draft No.1, §21) explores whether it should state a general rule concerning the availability of medical monitoring. Simultaneously, courts struggle with whether medical monitoring should ever be allowed in the context of prescription drugs or medical devices, which are far different from the environmental, toxic tort origin of this kind of claim. This post comments on the drug and device question, and the Restatement situation.
Background
Those jurisdictions recognizing medical monitoring general imposed some version of these elements: significant exposure; to a proven hazardous substance; through defendant’s tortious conduct; resulting in significantly greater risk of a serious latent disease; for which periodic diagnostic testing different from what would be proper in the absence of the exposure is reasonably necessary; and those tests make early detection of the disease possible. Of course, there is significant variation in the black-letter law among jurisdictions allowing medical monitoring claims. Another aspect of medical monitoring is the procedurally parasitic nature of the claim – it is almost never pursued, except via class action.
Drugs and Medical Devices
So how does this relate to drugs and medical devices? Although originally arising in environmental toxin cases, the medical monitoring notion has been applied to pharmaceuticals, with the drug as the “toxic” substance, voluntary ingestion by prescription the “exposure,” and potential future side effects as the increased risk of disease. An example is Thompson v. Pfizer Inc., No. H-05-1985 (S.D. Tex. filed June, 2005), the usual class action alleging that Viagra®, causes eventual blindness, and demanding “future medical monitoring after taking Viagra.” Implantable (and allegedly defective) medical devices have also been analogized to toxic substances. In Brennan et al. v. Guidant Corp., No. 1:05-cv-0827 (S.D. Ind. filed June, 2005), plaintiffs proposed a nationwide class of people who were implanted with defibrillators that allegedly short circuit. Among other relief, the case seeks medical monitoring.
But do these fit? The elements of a medical monitoring claim were fashioned for environmental tort actions, including the requirement of significant toxic exposure above background levels – something that makes no sense with respect to drugs or devices, or other things that just aren’t in the environment. The earliest, foundational cases make clear that the remedy was designed to deal with environmental exposures. Friends For All Children, Inc. v. Lockheed Aircraft Corp., 746 F.2d 816 (D.C. Cir. 1984), probably the first reported monitoring case, involved possible effects of sudden airplane cabin de-pressurization. In re Paoli Railroad Yard PCB Litig., 916 F.2d 829, 850 (3d Cir. 1990), involved alleged PCB exposure. Ayers v. Twp. of Jackson, 106 N.J. 557 (1987), arose from chemical contamination of plaintiffs’ water supply.
The policy justification for medical monitoring recovery was the nature of an environmental tort action made proving specific causation difficult, and there was no adequate governmental program for testing possible victims of environmental toxic exposures. These factors are likewise absent in most (non-environmental) product liability cases – especially those involving voluntary consumers of a prescription drug regulated by the FDA and prescribed by a licensed doctor legally responsible for weighing product risks and benefits. Proving and establishing causation in the environmental context, say a cancer case with numerous possible alternative causes and no way to establish exposure levels, is a far cry from FDA-approved drugs, extensively tested in clinical trials, and prescribed by physicians making individual medical evaluations and keeping precise prescription records. Exposure is knowing and voluntary, and dosage exposure easily established. The evidentiary gaps medical monitoring was designed to fill a gap just don’t exist in the drug and device context. Moreover, the existence of FDA oversight of product approval, and the publicity that normally surrounds drug withdrawals negate the policy rationale of the environmental cases.
The link between a polluter’s conduct and increased disease risks from pollution is direct, apparent, and logical. Prescription medical products are different, since the risk of bodily harm wasn’t created by the defendant. That’s because all prescription products carry risks – why they require a prescription in the first place. “Safe” does not mean risk-free, any more than “effective” means that every patient is cured. The FDA weighs the safety and efficacy of the drug in general, and the prescribing physician weighs individual risks and benefits. No exposure is often not an option. Rather, the alternative is frequently a competitive drug in the same class, carrying similar risks. Unless the risk of adverse effects is so widespread and serious – the Restatement Third, Products Liability test of risk so great that no reasonable doctor would prescribe for any class of patient – the mere risk of future bodily harm should not be actionable in this context.
The Restatement Question
The ALI in its Council Draft of the Restatement of the Law (Third) Economic Torts and Related Wrongs is currently exploring whether a general rule can be formulated concerning the availability of medical monitoring. Section 21 of this draft would recognize an action to recover monitoring expenses (1) to prevent or mitigate a risk of serious bodily harm when the defendant would be liable to the plaintiff if the bodily harm became manifest, (2) if the expense is required by a risk of bodily harm created by the defendant, (3) the expense provides no other material benefit to the plaintiff, (4) when the plaintiff has incurred or will incur the expense, and (5) if the liability is not “indeterminate.”
That’s badly premature – and it’s certainly not “restating” any general rule recognizing medical monitoring. Rather courts are badly split on even whether such a claim should exist. Actually, the trend over the last several years has been against the recognition of medical monitoring claims where there’s no present injury. The most recent case cited in the Reporter Note as supporting medical monitoring is seven years old, and the Supreme Court decision in Metro-North Commuter R. Co. v. Buckley, 521 U.S. 424 (1997), stands as an inexact but unmistakable turning point away from expanding the law. Since then most courts have passed on this type of claim. See Wood v. Wyeth-Ayerst Labs., 82 S.W.3d 849, 857 (Ky. 2002); Hinton v. Monsanto Co., 813 So.2d 827, 831 (Ala., 2001); Henry v. Dow Chemical Co., 473 Mich. 63, 701 N.W.2d 684 (Mich. 2005). Several federal courts have weighed in, also in opposition to letting uninjured people sue for medical monitoring. Norwood v. Raytheon Co., 414 F.Supp.2d 659 (W.D. Tex. 2006); Paz v. Brush Engineered Materials, Inc., 351 F.Supp.2d 580, 586 (S.D. Miss. 2005); but see Allgood v. General Motors Corp., 2006 WL 2669337 (S.D. Ind. Sep. 18, 2006).
Beyond the existential point, the doctrine is so new in most jurisdictions that appellate courts have yet to evolve workable elements based on practical experience drawn from real-world fact patterns. This immaturity shows up in the confusion over the wrongful conduct element of the claim. Some jurisdictions, such as Pennsylvania require causation by a defendant’s negligence; others, any type of “tortuous conduct” – a term broad enough to embrace faultless conduct, such strict liability and many consumer fraud act claims. Given the extraordinary nature of the relief – payments to uninjured people – any Restatement rule should be limited to negligence or intentional fault.
Another unsettled area is causation – not surprising, since there’s no present injury. Does the defendant’s conduct need to cause the exposure, or the increased risk, or both? Must the defendant’s conduct cause the need for medical monitoring? Plaintiffs predictably argue that any point in the causal chain, no matter how remote, is enough. Their preferred causal chain runs: (1) defendant commits some unreasonable act regarding a product; (2) plaintiff uses the product; (3) product use entails a risk of future harm. This amounts to negligence “in the air”: conduct related only to the product, but not to either the plaintiffs’ use of the product or the alleged need for medical monitoring. It is enough, they assert, that use of the product caused their risk.
The current Restatement Draft is not altogether clear, but apparently requires that the a risk of harm be created by the defendant. Mere sale of a product as to which some aspect of the defendant’s design, manufacture, or marketing can be criticized is not creation of risk. Rather, risk is “created” when the defendant’s wrongful conduct causes the risk for which the plaintiff seeks monitoring for, both by creating (or negligently permitting) the hazard of the product and causing plaintiff to be exposed to the risk improperly. This is important in the drug/device context. When a company produces, say, a drug that has significant medical benefit – but also real and definite hazards (potential side effects), it is usually theoretically possible to monitor those risks. But the FDA must approve the drug on a risk/benefit analysis. Who “created” the risk? The company that submitted the drug for FDA approval, or the FDA in approving it as more beneficial than risky? The risk is known, and has been accepted by the government as reasonable under the circumstances. Later, suppose there is wrongful conduct in marketing, say a misrepresentation as to efficacy – but the plaintiff (or prescribing physician) did not rely on that representation and voluntarily encounters the risk. Once again causation is absent. The conduct may have created an abstract risk, but did not create risk to this plaintiff. It ought to be clear that this is not “creation of the risk.”
“Risk” also needs clarification. Even jurisdictions recognizing medical monitoring that do not require the future harm be likely or certain typically require more than a bare increased risk. Instead there must be a significant risk, an increased risk, a medically significant risk, or similar formulation. The Restatement draft needs to be clarified not to sanction litigation over de minimis, insignificant, or essentially background level risks. See O’Neal v. Department of the Army, 852 F.Supp. 327, 336 (M.D. Pa. 1994). Also, it should be clarified that the requisite “risk” – whatever levels it may be – is the risk to an individual plaintiff created by the harmful exposure above and beyond the background risk for that injury in the relevant population and above and beyond that created by plaintiff-specific risk factors from other causes. To some degree “over and above” is reflected in the requirement that the monitoring provide no other material benefit to the plaintiff. But it is more than that. Plaintiffs typically argue that monitoring need only be different from testing appropriate for the population at large, and that “risk” means no more than use of a product increasing some risk in some of the people using the product. But such generic increased risk should not be enough do support recovery without present injury. It’s the “in the air” problem once again. Plaintiffs must do more than establish that a product generally causes an increased risk in some users. Rather, the requirement should be that the product sufficiently increased the risk of future disease in each particular plaintiff seeking medical monitoring. This is a crucial distinction, since specific causation has traditionally been the standard in tort. In re Prempro Products Liability Litigation, 230 F.R.D. 555 (E.D. Ark. 2005).
The Restatement draft requires that the “expense is required by a risk of bodily harm.” For the expense to be “required” by the risk, there must be a scientific and medical underpinning to the monitoring regime requested. States allowing monitoring typically require that the monitoring be reasonably medically necessary. Thus monitoring must have more than speculative medical value. Public health experts also routinely weigh the necessity, risks, and benefits of medical monitoring programs. Critical factors include whether monitoring can detect – reliably, accurately, and consistently – the disease earlier than it would otherwise be detected; whether early detection can improve mortality – as opposed to other less precise measures of outcome; whether available diagnostic testing itself has risks – either because it is invasive or because it has a high rate of “false positive” results that might lead to invasive follow-up testing; and economic costs. Medical monitoring proposals created for litigation may be touted as “required” by plaintiffs’ experts, while containing testing not prescribed routinely by physicians in actual practice, and/or studied and rejected by the public health community. The danger here is creation of a cause of action for the early detection of disease that is at odds with the medical community’s view on prevention. Plaintiffs should not be allowed to urge “required” medical testing that is not routinely prescribed – and may be expressly rejected – by the medical standard of care.
A final point on the Restatement: §21(d) of the Draft requires that liability not be “indeterminate.” If that means relief cannot be class-based, then it’s a good thing. Liability is indeterminate when a potential liability is so uncertain in time, class, or amount that an actor cannot fairly or practically be expected to account for the potential liability in determining the conduct giving rise to the potential liability. As ordinarily demanded, medical monitoring virtually always risks indeterminate, class-wide liability. That’s why, as a practical matter, the claim uniquely does not exist outside the class action context; very few individual medical monitoring claims have ever been brought. In the class context, liability virtually always is unlimited and unpredictable. Until the final stage of the litigation, class size is merely estimated for the limited purpose of assessing numerosity. Actual interest of absent class members in participating in the testing is assumed without empirical basis. The peculiarly future-oriented nature of the relief – unconstrained by present injury – begs several questions. What is the effect of a change in medical technology? If testing is approved by the jury as having met the appropriate standard, but is later shown not to work (as occurred with chest x-rays for detecting lung cancers), how is the monitoring halted? If medicine evolves a new test that appears to work better, can plaintiffs sue again? Can a court somehow modify an existing award of medical monitoring and order defendant to pay more for tests that were not the subject of the trial? What showing would plaintiffs need to make to get new testing? If existing programs – designed to last for decades given the latency periods of many diseases – are not modifiable to reflect changing state of the art, what good are they from a public health and tort policy perspective?