Prod.liab.rep. (cch) P 13,013mary Kathryn Thomas, Plaintiff-appellant, v. Hoffman-laroche, Inc., Defendant-appellee, 949 F.2d 806 (5th Cir. 1992)

Annotate this Case
US Court of Appeals for the Fifth Circuit - 949 F.2d 806 (5th Cir. 1992)

Jan. 3, 1992


John A. Owens, Susie T. Carver, Phelps, Owens, Jenkins, Gibson & Fowler, Tuscaloosa, Ala., for plaintiff-appellant.

Amy Johnson, Henry & Kelly, Austin, Tex., for amicus curiae, Public Citizen.

Earl Keyes, Keyes, Moss, Piazza & Woods, Jackson, Miss., Frederick T. Davis, John D. Winter, Karl E. Seib, Jr., Patterson, Belknap, Webb & Tyler, New York City, for defendant-appellee.

Appeal from the United States District Court for the Northern District of Mississippi.

Before WISDOM, KING and BARKSDALE, Circuit Judges.

WISDOM, Circuit Judge:


Mary Kathryn Thomas sued Hoffman-LaRoche, Inc., the manufacturer of a prescription acne medication called Accutane, under a failure to warn theory1  for injuries she sustained allegedly as a result of taking Accutane. The district court granted Hoffman-LaRoche's post-trial motion for judgment notwithstanding the verdict, setting aside a jury award of one million dollars. The district court held that Thomas had failed to introduce sufficient evidence to show that the allegedly inadequate warning caused her injury. 731 F. Supp. 224. We AFFIRM.

A. The Injury.

In late January 1984, the plaintiff had an appointment with her dermatologist, Dr. Robert P. Myers. She had been under Dr. Myers's treatment since 1979 for acne rosacea. When she met with Dr. Myers, Dr. Myers diagnosed Thomas as having developed a more serious form of acne, cystic acne. The cystic acne had proved resistant to the antibiotic treatment that Dr. Myers had prescribed for Thomas's acne rosacea, so Dr. Myers decided to prescribe Accutane for Thomas. At a monthly check-up on February 23rd, Thomas reported experiencing dry skin and headaches as side effects of the Accutane. Dr. Myers prescribed a skin cream for the dry skin, but, because of success in treating Thomas's acne, maintained Thomas's Accutane treatment.

Several days later Dr. Myers received a telephone call from Thomas's daughter who told Dr. Myers that her mother was disoriented and confused. Dr. Myers met with Thomas the following day, took her off Accutane, and referred her to a local internist, Dr. Dill. Dr. Dill, in turn, referred Thomas to Dr. Lawrence Mahalak, a neurologist. Thomas was hospitalized on February 29, 1984, suffering from a fever and disorientation.

On March 6th, Thomas was transferred to the Mayo Clinic in Rochester, Minnesota. While at the Mayo Clinic, March 12th, Thomas suffered her first seizure. She recovered from her fever and was discharged on March 23rd. On discharge, Dr. William Swanson, the treating physician at the Mayo Clinic during her March stay, wrote that the cause of her illness was "uncertain although the course [of her illness] would be most compatible with some type of infectious etiology, most likely viral."

On May 16th, Thomas was again admitted to a local hospital, under the care of Dr. Mahalak, suffering from a fever and disorientation. She was diagnosed as having pneumonia. While in the hospital, Thomas suffered her second seizure. She was transferred to the Mayo Clinic on May 22nd. At the time of her transfer, Dr. Mahalak, in consultation with Dr. William Causey, an infectious disease expert, attributed the seizures to a recurrent central nervous system infection. Thomas recovered and was released from the Mayo Clinic on May 26th. At the time of her discharge from the Mayo Clinic, the neurologist in charge of her treatment, Dr. J.D. Bartleson, stated that he was unable to reach a diagnosis.

In early June 1984, Thomas, again suffering from fever and disorientation, was admitted to the University of Southern Alabama Hospital under the care of Dr. William Kase, a board certified neurologist. She recovered from her symptoms and was released within four days.

In February of 1989, Thomas, again suffering from fever and disorientation, was admitted to a local hospital under the care of Dr. Mahalak. While in the hospital, she suffered a third seizure. Every expert who testified as to the cause of the 1989 seizure agreed that it was most likely caused by a virus, though a specific virus was neither isolated nor identified. The experts who expressed an opinion on the issue agreed that this third seizure was not related to the taking of Accutane in 1984.2 

At trial, Thomas introduced the testimony of three experts to establish that her use of Accutane had caused her 1984 seizures. Dr. O'Donnell,3  an expert with a Ph.D. in Pharmacy, testified that he believed that Accutane did cause seizures in some patients, though he was not permitted to express his opinion as to whether the use of Accutane caused Thomas's seizures. Two physicians, Dr. Mahalak and Dr. William J. Riley,4  testified that, to a reasonable degree of medical certainty, Accutane had caused Thomas's two 1984 seizures.

Hoffman-LaRoche introduced Thomas's contemporaneous medical records, and the testimony of four other physicians, Dr. Causey, Dr. Kase, Dr. Swanson, and Dr. Elias Chalhub.5  The records and the testimony introduced by Hoffman-LaRoche consistently pointed to an infectious process as the most likely cause of Thomas's fevers and seizures.

Despite the fact that over a million patients had used Accutane at the time of trial, no epidemiological study was presented that would suggest a causal connection between Accutane and seizures. One epidemiological study was completed and available at the time of trial, but neither party introduced the study.

B. The Warning Evidence.

Before Dr. Myers prescribed the drug, Hoffman-LaRoche had made available to him information that the product had been effective in treating patients with acne, but warned that a number of side effects were common. With respect to Accutane, the 1983 version of the Physician's Desk Reference ("PDR") stated:

Because of significant adverse effects associated with its use, Accutane should be reserved for patients with severe cystic acne who are unresponsive to conventional therapy, including systemic antibiotics.

In addition to this general warning, the PDR description also warned of a number of specific side effects that had been observed in Accutane patients: cheilitis, conjunctivitis, musculoskeletal symptoms, rash, temporary thinning of hair, skin peeling, skin infections, nonspecific urogenital findings, nonspecific gastrointestinal symptoms, fatigue, headache, and increased susceptibility to sunburn. The 1983 PDR description noted that use of Accutane was contraindicated for pregnant women and for patients who are sensitive to parabens.6  By January 1984, Hoffman-LaRoche had also provided specific warnings by letter concerning three other side effects observed in Accutane patients: pseudotumor cerebri, corneal opacities, and skeletal hyperostosis.7  The company provided the same information in inserts in the packages containing the product. The plaintiff seeks recovery arguing that Accutane was unreasonably dangerous despite these extensive warnings because Hoffman-LaRoche failed to list seizures as one of the potential side effects.

The testimony at trial established that, at the time Dr. Myers prescribed Accutane for the plaintiff's acne condition, doctors had prescribed Accutane to some four hundred thousand patients, of whom nine had reported "seizures"8  after using Accutane. A tenth seizure was reported to Hoffman-LaRoche in late January of 1984, several days after the plaintiff began her Accutane treatment. The evidence further established that, by May of 1989, over a million patients had been treated with Accutane. Forty-nine seizures had been reported.

Hoffman-LaRoche received these reports and forwarded them to the Food and Drug Administration. Because it was unclear whether these seizures were caused by the use of Accutane, Hoffman-LaRoche did not specifically mention seizures in its warning at the time Dr. Myers prescribed Accutane for the plaintiff. Hoffman-LaRoche added a reference to seizures in the 1985 PDR, and sent a "Dear Doctor" letter9  in the spring of 1984, notifying physicians of the change in the PDR description. Even in the spring of 1984, it remained unclear whether these incidents were caused by the use of Accutane; as a result, Hoffman-LaRoche's warning stated only that seizures "ha [d] been reported in less than one percent of patients and may bear no relation to treatment".

At trial, Thomas presented the testimony of Dr. O'Donnell to support her position that the warning accompanying Accutane was inadequate and that an adequate warning would have prevented her injury. Dr. O'Donnell testified that the warning given by Hoffman-LaRoche was inadequate for two reasons. First, Hoffman-LaRoche should have warned of the possibility of a slight risk of seizures associated with the use of Accutane in the fall of 1983, instead of in the spring of 1984. Second, because of the overall level of risk associated with Accutane, the warning should have stated that use of the product is contraindicated for all patients other than those with severe cystic acne.10 

With regard to whether an adequate warning would have changed Dr. Myers's decision to use Accutane, Dr. O'Donnell first testified:

In my opinion, if the warnings were complete as I described them and the use in less than severe conditions of acne was contraindicated, a reasonable physician would not have prescribed [Accutane] for Mrs. Thomas.

Counsel for Hoffman-LaRoche objected, on the basis that the opinion assumed a diagnosis of Thomas's condition that was not supported by any credible evidence.11  The trial judge sustained the objection, but, after discussion, permitted Dr. O'Donnell to testify on the issue, provided that he not refer to a diagnosis unsupported by the evidence. Dr. O'Donnell did so:

Q. Doctor, do you have an opinion as to what a reasonably knowledgeable physician, as to whether or not a reasonably knowledgeable physician who was given the warnings that you have told the jury you thought should have been in effect for Accutane in January of 1984, whether a reasonable physician so warned would have prescribed Accutane to Mrs. Thomas?

A. I do have an opinion, sir.

Q. And what is that opinion?

A. A reasonable physician would not have prescribed the Accutane for Mrs. Thomas.

Dr. Myers, on the other hand, testified that he had read in the fall of 1983 of the possibility that Accutane may have caused seizures in very rare cases. He further testified that he would not have changed his decision if warned of the possible risk of seizure.

C. Resolution of the Issues at Trial.

The case was submitted to the jury, and the jury returned a judgment in favor of Thomas in the amount of one million dollars. Hoffman-LaRoche moved for a judgment notwithstanding the verdict, or alternatively for a new trial. In resolving the motion, the trial judge held that a plaintiff must prove two facts to establish that a failure to warn of a risk associated with a prescription drug caused the plaintiff's injury. First, he held that a plaintiff must demonstrate that an adequate warning would have changed the treating physician's decision to prescribe the drug for the plaintiff. Second, he held that a plaintiff must also show that the drug caused the plaintiff's injuries. The trial judge held that there was insufficient evidence to support a finding in favor of the plaintiff, and granted Hoffman-LaRoche's motion for judgment notwithstanding the verdict.

Under Mississippi law,12  a product may be unreasonably dangerous if the manufacturer fails to warn of a non-obvious risk associated with the normal use of the product. When the product in question is a prescription drug, such as Accutane, Mississippi follows the learned intermediary doctrine.13  Under this doctrine, the manufacturer's failure to warn the patient of the product's risks does not render the product defective or unreasonably dangerous so long as the manufacturer adequately warns the learned intermediary.14  Both parties agree that Thomas bears the full burden of establishing that Accutane was the cause of her 1984 seizures.15  They disagree over whether Mississippi law requires the plaintiff to establish that an adequate warning would have changed Dr. Myers's decision to prescribe Accutane for Thomas.

In resolving this dispute, we begin with the Mississippi Supreme Court's recent decision in Wyeth Laboratories, Inc. v. Fortenberry.16 

A. Inadequate Warning and Causation.

In Fortenberry, the plaintiff brought suit, against his doctor and Wyeth Laboratories, alleging injuries caused by a non-swine influenza vaccine. The plaintiff alleged that the product was unreasonably dangerous because of an inadequate warning.17  The Fortenberry Court rejected his claim, finding that the warning was adequate as a matter of law.18  The Fortenberry Court also addressed whether a plaintiff must establish causation:

Assuming arguendo that the warning was inadequate, Fortenberry still had the burden of showing that an adequate warning would have altered Dr. Moore's conduct.19 

Thomas characterizes this statement as dictum and urges us to ignore it. We cannot do so. Under diversity jurisdiction, considered dictum stating how an issue is to be resolved is evidence we must consider in determining the most likely result to be reached by a Mississippi court faced with this issue.20  Given that no other Mississippi court has spoken on the subject, we consider this plain and unconditioned statement by the Fortenberry Court to be convincing evidence of the likely result under Mississippi law. We, therefore, hold that, under Mississippi law, in a prescription drug failure to warn case, the plaintiff must establish that an adequate warning would have convinced the treating physician not to prescribe the product for the plaintiff.21 

To satisfy the burden of establishing warning causation, a plaintiff may introduce either objective evidence of how a reasonable physician would have responded to an adequate warning,22  or subjective evidence of how the treating physician would have responded.23  But, to create a jury question, the evidence introduced must be of sufficient weight to establish, by the preponderance of the evidence, at least some reasonable likelihood that an adequate warning would have prevented the plaintiff from receiving the drug.

B. A Presumption of Causation?

Accepting that she must prove that an adequate warning would have prevented her from receiving the drug, Thomas asks us to presume warning causation if a warning is inadequate, and cites in support of her presumption comments j and k of section 402A of the Restatement (Second) of Torts. In relevant part, comment j provides:

Where warning is given, the seller may reasonably assume that it will be read and heeded; and a product bearing such a warning, which is safe for use if it is followed, is not in defective condition, nor is it unreasonably dangerous.24 

In 1972, the Texas Supreme Court suggested in dicta that comment j could be read to create a rebuttable presumption in favor of a plaintiff that the plaintiff would have read and heeded a warning if it had been given.25  Building on this dicta, we predicted, in Reyes v. Wyeth Laboratories,26  that Texas courts would adopt the following presumption:

Where a consumer, whose injury the manufacturer should have reasonably foreseen, is injured by a product sold without a required warning, a rebuttable presumption will arise that the consumer would have read any warning provided by the manufacturer, and acted so as to minimize the risks.27 

No Mississippi court has adopted a presumption of causation on facts similar to those in this case, and the failure of the Mississippi Supreme Court to mention such a presumption in Fortenberry strongly suggests that such a presumption does not exist under Mississippi law.

Furthermore, in attempting to predict the likely course that Mississippi law will take, we recognize that there are two very different types of warnings that might be associated with a particular product: (1) an unavoidable risk warning; and (2) a preventable risk warning. The first type of warning details a risk that a consumer28  cannot avoid if the consumer chooses to use the product. Typically, warnings associated with medical drugs fall into this category. For example, a product description may warn that one in four hundred thousand patients will suffer a certain adverse reaction. There is nothing that can be done about the risk, and often no way to discern in advance who will be the one person in four hundred thousand to experience the adverse reaction. The only question for the potential consumer of the product is whether the potential benefit to be obtained from using the product outweighs this risk. The second type of warning is customarily associated with mechanical products, and details risks that can be avoided by using the product in a certain manner.29  For example, a pressurized can may warn users of the dangers of puncturing the can or of exposing the can to extremes of temperature. By heeding the warning, the consumer may use the product safely and avoid the detailed risks.

The Reyes presumption provides that a product user will read a warning and act to minimize his level of risk. Typically, the choice facing the user in the preventable risk situation is between using the product safely and using the product unsafely. Because the precautions are typically minimal (i.e., store the pressurized can away from the water heater), we have little trouble with a rebuttable presumption that a reasonable product user will choose to use the product safely. The choice, however, presented by an unavoidable risk warning is not between the safe use and the unsafe use of a product, but between using and not using the product. The consumer can choose to use the product and face its risks, or choose not to use the product and lose its potential benefits. Generally, using the product will present the less risky of these two alternatives. Consider the polio vaccine in Reyes. There is a risk of contracting polio if a person uses the vaccine, and there is a risk if the person decides not to use the vaccine. Presumably, the risk of contracting polio is less with the vaccine than without it. If we assume that the average user will take the less risky alternative, the average user will choose to take the vaccine, and a warning detailing the unavoidable risks of the polio vaccine, whether given or not, would not change that decision.30  Unless the plaintiff can establish that using the product is, for the average consumer, the more risky alternative, the Reyes rule that the consumer will act to minimize his level of risk, if applied in the context of an unavoidable risk, would seem to establish a rebuttable presumption that the consumer would not have changed his decision to use the product if warned of the unavoidable risk.

For this reason, we reject Thomas's contention that causation is presumed, given an inadequate warning, in the context of an unavoidable risk. We are willing to assume that the failure to give an adequate warning of a known risk entitles the plaintiff to a rebuttable presumption that the learned intermediary would have read and heeded a proper warning. But "heed" in this context means only that the learned intermediary would have incorporated the "additional"31  risk into his decisional calculus.32  The burden remains on the plaintiff to demonstrate that the additional non-disclosed risk was sufficiently high that it would have changed the treating physician's decision to prescribe the product for the plaintiff.33 

In summary, in a failure to warn case involving a prescription drug, Mississippi law requires a plaintiff to establish that an adequate warning would have prevented the plaintiff's injury. To establish that element, the plaintiff must establish, by the preponderance of the evidence, both: (1) that an adequate warning would have prevented the treating physician from administering the drug; and (2) that the injury would not have occurred had the drug not been administered.

In reviewing a judgment notwithstanding the verdict, we view the evidence in the light most favorable to the non-moving party.34  If, in that light, a reasonable jury cannot find an essential element of the plaintiff's case, then we will affirm the judgment notwithstanding the verdict in favor of the defendant.35 

Warning Causation: Sufficiency of the Evidence.

In his testimony, Dr. O'Donnell testified that the actual warning given was inadequate, and rendered Accutane unreasonably dangerous, for two reasons: (1) Hoffman-LaRoche failed to send a "Dear Doctor" letter to all physicians detailing the available information on the reported seizures in the fall of 1983; and (2) Hoffman-LaRoche failed to contraindicate use of Accutane for all patients except those suffering from severe cystic acne.36  Before we address the warning causation issue, we consider whether the evidence in the record will support a reasonable jury finding that Accutane was unreasonably dangerous in the absence of these two additional warnings.

Mississippi law requires a manufacturer to give a reasonable warning.37  To be reasonable, the warning should neither understate nor overstate the known risks associated with the use of a particular product. We consider whether a reasonable jury could have found either aspect of the warning recommended by Dr. O'Donnell to be reasonable under the circumstances of this case.

The undisputed evidence establishes that, by the end of January 1984, ten patients had reported seizures during their treatment with Accutane. At trial, all of the plaintiff's experts agreed that the seizures experienced by nine of the ten patients were eventually determined to be unrelated to the taking of Accutane, and that the tenth most likely involved a fainting spell.38  (It remained uncertain whether the tenth reported incident was caused by the taking of Accutane.) The evidence also suggested that the Accutane patients had fewer seizures than would have been expected, on average, due to the background risk of seizures in the general population.

Based on the available evidence, a reasonable jury could determine that there was at least the possibility that seizures were caused by Accutane in very rare cases. We assume, but do not decide, that a reasonable jury could have found the warning accompanying Accutane to be inadequate for failing to provide information on this possibility.

With regard to the reasonableness of Dr. O'Donnell's recommended contraindication, we find insufficient evidence to support a finding that Accutane was unreasonably dangerous absent such contraindication. As Dr. O'Donnell and several other expert witnesses testified, a statement that a particular use is contraindicated does much more than a typical warning. Where a typical warning lists possible side effects and their likelihood, permitting the treating physician to balance the potential costs and benefits in a particular situation, a contraindication is essentially a direction to the physician not to use the drug in certain circumstances. When a use is contraindicated, the manufacturer and the FDA have already balanced the costs and benefits of the drug and determined that, under the stated conditions, the risks of using the product will always outweigh the benefits. As all of the witnesses agreed, a physician will not prescribe a product when its use is contraindicated, absent exceptional circumstances.

In order to justify a particular contraindication, a plaintiff must introduce evidence that the risks of the drug will almost always outweigh its benefits in the situation contraindicated. The only evidence that the risks of Accutane outweighed its benefits for all patients except those with severe cystic acne comes from Dr. O'Donnell. He testified that the use of Accutane should have been contraindicated in all cases other than severe cystic acne. In essence, he is testifying that the risks of Accutane outweigh its benefits for everyone other than a patient with severe cystic acne. Two undisputed facts stand in sharp contrast to Dr. O'Donnell's opinion.

First, the FDA, after considering all of the evidence, provided a contraindication only for "patients who are pregnant or who intend to become pregnant while undergoing treatment" and for "patients who are sensitive to parabens". Despite knowledge of the risks associated with Accutane, the FDA did not consider it so risky as to require a contraindication for all patients except those with severe cystic acne. Instead, the FDA decided to leave that decision to the individual physician in each particular case. Second, the undisputed testimony at trial established that over one million patients had used Accutane, and that, at most, only 360,000 persons in the United States suffer from cystic acne.39  Thus, despite the risks of Accutane, it is undisputed that thousands of physicians prescribed Accutane for patients who did not suffer from severe cystic acne.

In resolving this issue, we cannot avoid the fact that the products liability system plays a dual role of compensating injured victims and regulating products. If we permit the subjective opinion of a single expert to create a genuine issue of material fact in the face of the decisions of the FDA and thousands of physicians so that a particular plaintiff can recover, we would be denying access to Accutane for hundreds of thousands of acne sufferers.40 

In Matsushita Electric Industrial Co. v. Zenith Radio Corp.,41  the Supreme Court addressed a similar issue. In Matsushita, the Court held that there was insufficient evidence to create a jury issue on the existence of a predatory pricing conspiracy despite expert testimony that such a conspiracy could and did exist.42  Because the undisputed historical evidence from the market established that such a conspiracy would be unprofitable, the Court held the expert testimony insufficient to create a jury issue.43  The Court feared that to permit the case to go to the jury in the face of the market evidence would discourage competition,44  either because the actual jury might reach an unreasonable result, or because of the high cost of convincing the jury to reach the right answer.45 

We find this analysis fully applicable to this case. Thousands of physicians, and hundreds of thousands of patients, have judged Accutane to be worth the risk, even after the possible seizure risk was disclosed. Because this balancing decision is inescapably a personal one,46  the opinion of Dr. O'Donnell that Accutane is not worth the risk is insufficient to create a genuine issue of fact in the face of these informed decisions. We, therefore, hold, as a matter of law, that the failure to include the recommended contraindication did not render the warning inadequate.

Assuming that the failure to inform physicians of the possibility of seizures in the fall of 1983 rendered Accutane unreasonably dangerous, we now consider whether Thomas has introduced sufficient evidence to permit a reasonable jury to find that an adequate warning would have prevented Thomas's injuries. Dr. O'Donnell expressed no opinion on whether a warning concerning the possibility of a seizure risk, in the absence of his recommended contraindication, would have changed the decision of a reasonable physician to prescribe Accutane for Thomas. The careful wording of his first attempt at the warning causation issue strongly suggests that his opinion was based on the contraindication aspect of his recommended warning. Without the contraindication included in the warning, we cannot be certain that Dr. O'Donnell would have expressed the same opinion on the warning causation issue. We are not overly concerned, though, with what opinion Dr. O'Donnell might have expressed on this issue because, in examining the remainder of the record, we find evidence that eliminates any reasonable possibility that the release of the information concerning the ten reported seizures would have changed the decision of Dr. Myers to prescribe Accutane in this case.

It is undisputed that Thomas suffered from a serious acne condition. She had been under treatment for her condition for over seven years. While it is probable, though not entirely certain, that she was suffering from cystic acne when she went to Dr. Myers in January of 1984, it is certain that her acne was not responding to conventional therapy, including systemic antibiotics. Furthermore, there was nothing in Thomas's prior history to suggest that she would be the one person in so many thousands to suffer an adverse reaction to Accutane. Thus, Thomas was a reasonable, though perhaps not ideal, choice for treatment with Accutane.

In the spring of 1984, Hoffman-LaRoche disclosed the available information on the reported seizures to physicians. Yet, even after the release of this information, Accutane has remained a popular choice for acne treatment. Indeed, it appears from the record that the release of the seizure information had no effect on the overall level of Accutane use.47  Since the seizure warning in 1984, physicians have prescribed Accutane for hundreds of thousands of patients, the vast majority of whom were, like Thomas, reasonable, though perhaps not ideal, candidates for Accutane treatment. Furthermore, Thomas has failed to introduce evidence of even a single physician who actually changed his or her practice concerning Accutane based on the release of the seizure information.

We hold that this undisputed historical evidence concerning the usage of Accutane after an adequate seizure warning was given precludes any reasonable possibility that information on the ten reported seizures would have prevented Thomas's injury. The possibility that Dr. Myers would have changed his decision if he had been warned of the possibility of rare seizures caused by Accutane is too remote to create a genuine issue of fact with respect to warning causation.48  Because of the failure of the plaintiff to establish causation, we must affirm the judgment notwithstanding the verdict.49 IV. CONCLUSION

We hold that under Mississippi law a plaintiff in a prescription drug failure to warn case must establish, by a preponderance of the evidence, that the inadequate warning caused the plaintiff's injuries. The plaintiff must show that an adequate warning would have convinced the prescribing physician not to prescribe the drug for the plaintiff. No reasonable jury could conclude that Dr. Myers would not have prescribed Accutane for Thomas. We AFFIRM the judgment notwithstanding the verdict granted by the district court.

Because of our affirmance of the district court's decision as to the adequacy of the warning, we need not address the drug causation issue.

 1

Thomas alleged both that Hoffman-LaRoche was negligent for failing to warn and that the product Accutane was unreasonably dangerous because of the inadequacy of the warning. We consider both the negligence and the strict liability action together. Thomas also alleged a breach of express warranty, which issue is not before us on appeal

 2

Thomas contends that Accutane caused her to develop a seizure disorder and that she would suffer recurrent seizures because of her month-long treatment with Accutane. There is no competent evidence in the record to establish that Accutane could cause such a disorder

 3

Dr. O'Donnell was not involved in the treatment of Thomas, but was hired by her attorneys in anticipation of litigation. He was qualified, by the district court, as an expert in the field of medical warnings. Dr. O'Donnell testified that he was receiving between ten and fifteen thousand dollars for his testimony in this case, and that he was providing similar testimony in twelve other cases

 4

Dr. Riley also was hired in anticipation of trial. He had never examined Thomas, and expressed his opinion based on a brief review of her medical record. At the time Dr. Riley expressed his opinion, he did not have available information on the 1989 seizure. Several of the experts testified that the 1989 seizure was an important consideration in determining the likely cause of the 1984 seizures

 5

Dr. Chalhub was not involved in the treatment of Thomas, but was hired in anticipation of trial

 6

Parabens are used as a preservative in Accutane

 7

Some of these side effects had been observed only when Accutane was used to treat a patient with a disorder of keratinization, and had not been observed when Accutane was used to treat acne patients

 8

The "seizure" reports ranged from incidents that resembled fainting to grand mal seizures

 9

When a drug company discovers new information concerning side effects or possible adverse reactions to one of its products, the company typically sends a letter to physicians relating the new information. This letter is referred to as a "Dear Doctor" letter

 10

Dr. O'Donnell further testified that the warning remained inadequate despite the seizure language added by Hoffman-LaRoche in the spring of 1984. He testified that the warning should have read: "seizures have been associated with [Accutane] and may or may not bear relationship to therapy". We find untenable his suggestion that any reasonable physician would read his "associated with" and "may or may not bear relationship" language different from Hoffman-LaRoche's "have been reported in" and "may bear no relationship" language. Both warnings substantively connote the same thing: seizures have been observed in patients during the Accutane treatment period, and it is uncertain whether the seizures were caused by Accutane or coincidental

 11

Dr. Myers testified, and Thomas's contemporaneous medical records reflect, that Thomas was suffering from cystic acne in January of 1984, when Dr. Myers prescribed Accutane. Dr. Myers testified that, in 1980, Thomas had a less severe form of acne--acne rosacea. Both Thomas and her daughter, testified that they did not know Thomas had cystic acne in 1984. Thomas also elicited testimony from Dr. Myers that one generally knows if one has cystic acne. From this, Thomas argues that a reasonable jury could have inferred that Thomas did not have cystic acne in 1984, but was suffering from acne rosacea

 12

Both parties agree that Mississippi substantive law governs our analysis in this diversity action

 13

Wyeth Labs., Inc. v. Fortenberry, 530 So. 2d 688, 691-92 (Miss.1988)

 14

530 So. 2d at 691-92

 15

The parties do disagree over whether the evidence was sufficient to support such a finding

 16

 530 So. 2d 688 (Miss.1988)

 17

530 So. 2d at 691

 18

530 So. 2d at 692-93

 19

530 So. 2d at 691

 20

See, e.g., Towne Realty, Inc. v. Safeco Ins. Co. of America, 854 F.2d 1264, 1269 n. 5 (11th Cir. 1988)

 21

Fortenberry, 530 So. 2d at 691; see also Phillips v. Hull, 516 So. 2d 488, 493 (Miss.1987); Willett v. Baxter Int'l, Inc., 929 F.2d 1094, 1098-99 (5th Cir. 1991); Plummer v. Lederle Labs., 819 F.2d 349, 358-59 (2d Cir.), cert. denied, 484 U.S. 898, 108 S. Ct. 232, 98 L. Ed. 2d 191 (1987); Kirsch v. Picker Int'l, Inc., 753 F.2d 670, 671 (8th Cir. 1985)

 22

See, e.g., Hermes v. Pfizer, Inc., 848 F.2d 66, 69-70 (5th Cir. 1988)

 23

See Willett, 929 F.2d at 1099; Stanback v. Parke, Davis & Co., 657 F.2d 642, 645 (4th Cir. 1981). The D.C. Circuit has required objective evidence to establish causation in a similar situation. See Hartke v. McKelway, 707 F.2d 1544, 1550-51 & n. 6 (D.C. Cir.), cert. denied, 464 U.S. 983, 104 S. Ct. 425, 78 L. Ed. 2d 360 (1983); Henderson v. Milobsky, 595 F.2d 654, 658-69 & n. 26 (D.C. Cir. 1978); Canterbury v. Spence, 464 F.2d 772, 791 (D.C. Cir.), cert. denied, 409 U.S. 1064, 93 S. Ct. 560, 34 L. Ed. 2d 518 (1972)

 24

The "read and heed" aspect of this comment seems intended to preclude a finding of causation against a defendant in a products liability action so long as an adequate warning is given. Of course, as such, the "read and heed" aspect is superfluous. If an adequate warning is given, the product is not defective, and resolution of causation becomes irrelevant to liability

 25

See Technical Chemical Co. v. Jacobs, 480 S.W.2d 602, 605 (Tex.1972)

 26

498 F.2d 1264 (5th Cir.), cert. denied, 419 U.S. 1096, 95 S. Ct. 687, 42 L. Ed. 2d 688 (1974)

 27

498 F.2d at 1281

 28

The term "consumer" refers to the person making the decision whether a product should be used or purchased. In the context of a prescription drug, the term includes the learned intermediary who almost invariably makes the decision whether to prescribe the drug

 29

See, e.g., Pavlides v. Galveston Yacht Basin, Inc., 727 F.2d 330, 338-39 (5th Cir. 1984)

 30

In Reyes, we assumed that there was something the plaintiff could have done to reduce the risk of contracting polio from the vaccine. Reyes, 498 F.2d at 1281-82 ("Perhaps this [warning] would have prevented her polio.")

 31

Because it is uncertain whether the seizures were coincidental or causally related to the taking of Accutane, it is uncertain whether there was, in fact, any additional risk to be disclosed. At best, there was the possibility of an additional risk

 32

While we recognize that we reached the opposite result in Reyes, we do not follow that decision for two reasons: (1) we decided that case under Texas law, and we are deciding this case under Mississippi law; and (2) despite the opportunity, Texas courts have not adopted the Reyes presumption in unavoidable risk cases. See Cooper v. Bowser, 610 S.W.2d 825, 832-33 (Tex.Civ.App.--Tyler 1980, no writ)

 33

Instead of addressing the issues of warning inadequacy and warning causation separately, we could have combined the inquiries by defining the duty of a drug manufacturer to warn as requiring a warning only of those risks sufficiently serious that they would affect the average physician's decision to use the product. See P. Keeton, D. Dobbs, R. Keeton, D. Owen, Prosser and Keeton on Torts § 42, at 274-75 (5th ed. 1984)

 34

Boeing Co. v. Shipman, 411 F.2d 365, 375 (5th Cir. 1969) (en banc)

 35

See, e.g., Brock v. Merrell Dow Pharmaceuticals, Inc., 874 F.2d 307, 308-09 (5th Cir.), modified on other grounds on rehearing, 884 F.2d 166 (5th Cir. 1989), cert. denied, --- U.S. ----, 110 S. Ct. 1511, 108 L. Ed. 2d 646 (1990)

 36

Because Accutane is chemically related to Vitamin A, Dr. O'Donnell testified that the warning was also inadequate because Hoffman-LaRoche failed to list all of the potential side effects of Vitamin A. Hoffman-LaRoche did state that Accutane was related to Vitamin A, but it did not list all the known potential side effects of Vitamin A. While the potential side effects of a related chemical may be important in the early stages of a drug, by January of 1984, over 400,000 patients had taken Accutane. With this many patients, listing the actual side effects reported by Accutane patients provides a much better indication of the risks associated with Accutane, than would a list of the side effects of Vitamin A. Indeed, listing the side effects of Vitamin A, even though they had not been observed in these 400,000 patients, would tend to overstate the risks associated with Accutane. For that reason, we hold that Accutane was not unreasonably dangerous, as a matter of law, in January of 1984, for failing to list those side effects of Vitamin A that had not been reported or observed in the Accutane patients

 37

Fortenberry, 530 So. 2d at 692

 38

Of the ten reported seizures, three involved patients with a prior history of seizures, three involved patients who were diagnosed as having epilepsy unrelated to Accutane, one involved a patient who was eventually diagnosed as having a pre-existing seizure disorder, and two involved patients whose physicians specifically determined that the seizures were not related to Accutane. The tenth report involved a fainting spell, and no cause was determined

 39

It is unclear how many of the 360,000 suffered from severe cystic acne

 40

Alternatively, and perhaps 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

 41

 475 U.S. 574, 106 S. Ct. 1348, 89 L. Ed. 2d 538 (1986)

 42

475 U.S. at 593-95 & n. 19, 106 S. Ct. at 1359-60 n. 19

 43

475 U.S. at 588-93 & 595-97, 106 S. Ct. at 1356-59 & 1360-61

 44

475 U.S. at 593-94, 106 S. Ct. at 1359-60; see also Monsanto Co. v. Spray-Rite Serv. Corp., 465 U.S. 752, 762-64, 104 S. Ct. 1464, 1470-71, 79 L. Ed. 2d 775 (1984)

 45

As a third possibility, the Court might have been recognizing that most litigation is settled before a jury actually decides the case. Because of the uncertainty inherent in a jury verdict, permitting a case to go to a jury will substantially increase the settlement value of a case, even if, more often than not, the average jury will decide the case correctly. By taking the case from the jury, the Court removes that uncertainty and encourages conduct that it sees as likely to encourage, rather than discourage, competition

 46

It is simply not possible to compare in any meaningful manner the pain suffered by those few who experience an adverse reaction with the joy of the many who are cured of their acne. Whether the possibility of a cure is worth the risk of an adverse reaction is a balancing decision most appropriately left to the individual patient and the treating physician. When the relevant risks and benefits are disclosed, these two are best placed to decide whether the gamble of a particular treatment is worth the risk for a particular patient

 47

This is not surprising given that the risk of seizures for those taking Accutane was, from our perspective, indistinguishable from the background risk of seizures for the general population

 48

Cf. Willett, 929 F.2d at 1099

 49

See Willett, 929 F.2d at 1099; Plummer v. Lederle Labs., 819 F.2d at 358-59; Kirsch v. Picker Int'l, Inc., 753 F.2d at 671; Stanback v. Parke, Davis & Co., 657 F.2d at 645