Based in large part upon animal experimentation conducted by a scientist, Steve Carson, who was later convicted of fraud on the FDA, and upon admittedly misrepresented evaluations by Dow Corning's own Silas Braley, silicone gel became a popular substitute for breast tissue in the 1970's. Enclosed in a balloon constructed of amorphous silica and PDMS (polydimethalsiloxane), silicone gel breast implants were marketed and sold as safe, lifetime devices.
Lawyers for implant manufacturers now assert that silicone was tested and evaluated more thoroughly than any other product. The facts are, however, that no clinical trials were conducted prior to marketing, and only one such trial was in progress as late as 1983. That trial, the Gregory Study, had no controls and its only reported finding, which was never made public, was a 40% re-operation rate. After discovering the friability of the product, manufacturers began communicating this defect to operating surgeons in vaguely worded cautions contained in package inserts. I suspect that most of these surgeons never even knew the product contained a package insert. Indeed, former professor Hans Norberg, M.D., so testified. But for reasons best known to Med-Mal Insurers, doctors have uniformly refused to admit any such lack of knowledge. Nevertheless, this mode of communication (via package inserts) has been labeled even by the manufacturers as ineffective and inadequate. Sanders for BMS and Hayes for Dow concluded that there was not "a chance in hell" that a surgeon would read a package insert before surgery.
Did manufacturers represent that silicone gel breast implants would last a lifetime? Here are the facts.
- Dow Corning: In 1976 Dow Corning provides doctors a printed handout for patients which says that the product is expected to last a lifetime.
- BMS: As late as 1991, BMS was asserting publicly that its implants lasted a lifetime.
- Heyer-Schulte: In 1973, Heyer-Schulte stated that only its elastomer's outside surface ever came in contact with body tissue.
- 3M: In its training manual 3M taught its salesman that implants lasted a lifetime.
- Natural Y: Claimed in 1982 that its implants did not rupture or even bleed. BMS reiterated this claim for this poly product after the 1987 acquisition from Cooper.
Jan Varner, formerly at Dow, Heyer-Schulte, McGhan, 3M and INAMED testified that until the mid eighties silicone implants were marketed and sold as lifetime devices. Coyne, currently the MAN at 3M, in an early deposition said the same thing. The facts are, of course, that implants wear out over time. Average life expectancy for an intact implant is probably 10 (ten) years.
FRAUD
There seems little question therefore that every manufacturer engaged in fraudulent practices. As Judge Fawsett found in a recent trial against BMS and MEC in Florida: "MEC misrepresented its product through advertising and marketing ***." Barrow v. BMS, et al, 1998 WL 812318, 812345 (U.S.D.C. M.D. Fla. 1998). "Defendant intended to induce reliance upon such misrepresentation" Id at 812345.
"As a consequence of justifiable reliance on the misrepresentations and omissions of Defendant, Plaintiff suffered damages. Relying upon the misinformation given to her by her doctor, Plaintiff chose to be implanted with MEC's silicone gel breast implants. As a consequence, Plaintiff suffered a chronic foreign body reaction to the leaking silicone gel. She had inflammation of tissue in her breast area. After implantation Plaintiff suffered severe capsular contracture, on numerous occasions she had to undergo closed capsulotomies, and one of the Plaintiff's implants became deformed and produced the appearance of a bulge on her breast. She also suffered continuous foreign body reaction and chronic inflammation. She had an explant operation performed. Silicone gel remained in her breast tissue area and in her lymph nodes even after explantation. A significant amount of tissue was removed from Plaintiff upon explantation, and she has permanent scars on her breasts from removal of the implants, the capsular contracture, and adjacent breast tissue. She also suffers numbness in her breasts after explantation. Accordingly, Plaintiff is owed monetary damages to compensate for such injuries." Ibid.
Barrow was recently affirmed by the Eleventh Circuit, without opinion. BMS v. Barrow, 190 F. 3d 541 (11th Cir. 1999)
FRAUD BY PROXY
Lawyers for these manufacturers argue, however, that their clients made no misrepresentation directly to the patient and, thus, cannot be guilty of fraud. The argument, I believe, is specious for at least two reasons.
- First, the representations made by the manufacturer to the doctor were certainly intended to be passed on to the patient as a sales tool.
- Second, the doctor was clearly the agent of the manufacturer in making these representations to the patient although some courts characterize the doctor as the agent for the patient.
In Barrow, Judge Fawsett states with absolute clarity: "The fact that MEC made the misrepresentations, and omitted material information in its representations, to plaintiff's physician and not directly to plaintiff does not preclude recovery by plaintiff against MEC for such misrepresentations and omissions." Barrow 1998 WL at 812345. The court cites Albertson v. Richardson - Merrell, Inc., 441 So. 2d 1146 (Fla. Dist. Ct. App. 1983), in support of its conclusions, and Albertson is directly in point. There a "detail man" represented the safety of the product (Bendectin) to the doctor, not the patient, and the court stated that the manufacturer and the detail man could be held liable since the purpose of the misrepresentations was to encourage the physician to prescribe the drug for his pregnant patients. Id at 1150. The Florida court in Albertson, in turn, cited with approval, Wechsler v. Hoffman - La Roche, 99 N.Y.S. 2d 588 (Sup. Ct. 1950), affirmed as modified, 108 N.Y.S. 2d 990 (1951). In that case the court adopted two reasons for imposing liability. First, the court said as a matter of policy the manufacturer should be liable in fraud. Second, the court said that liability would be imposed because the doctor was acting for the patient and, therefore, a fraud upon the doctor was a fraud upon the patient. Id at 1150. The Albertson court also relied upon two Texas cases in imposing liability for fraud. Crocker v. Winthrop Laboratories, 514 S.W. 2d 429 (Tex. 1974); Bristol-Myers Co. v. Gonzales, 561 S.W. 2d 801 (Tex. 1978). Crocker, however, involved only product liability under 402A and indirect misrepresentations under 402B. Gonzales, is the same. Nevertheless, the analogy between 402B and fraud is pretty compelling.
Indeed, as far back as 1890, the Supreme Court of Texas adopted the Commissioners' statements: "As a general proposition, it may be correct, as contended by appellant, that a misrepresentation made to one person, and not with a view of revealing another, can not be available to another who may have acted on it ***. But it is sound doctrine that a third person to whom they were not directly made, can maintain an action for deceit, and seek cancellation of the contract made by him, if it appears that the defendant's false representations were made with a direct intent that he should act upon them in the manner which occasioned the injury." Gainesville Nat. Bank v. Bamberger, 13 S.W. 959 (Tex. 1890). While this case is cited in Tex. Jr. 3rd, it does not appear in any reported case I can find where this particular issue was involved. The spirit of the case, however, can be found in Cook Consultants, Inc. v. Larson, 700 S.W. 2d 231 (Tex. Civ. App. - Dallas 1985), where a surveyor who negligently misrepresented certain boundaries was held liable to a down the line purchaser. The court's extensive analysis of this cause of action is certainly instructive. Id at 234 - 236.
Westcliff Co. v. Wall, 267 S.W. 2d 544 (Tex. 1954), does not hold to the contrary. There an eavesdropper brought suit for fraud, and no recovery was permitted. The court cited Cooley on Torts: "No one has a right to accept and rely upon the representations of others but those to influence whose actions they were made." Id at 546. By implication, at least, if the misrepresentations are made for the purpose of influencing a doctor and his patient to act upon them, a cause of action will stand.
On the other hand, the Dallas Court of Appeals quoted the same section of Cooley on Torts to find that a misrepresentation by an insured to obtain a life insurance policy was not a wrongful act as to an insurer which reinsured the policy after the first company was put in receivership. First State Life Co. v. Stroud, 120 S.W. 2d 491 (Tex. Civ. App. - Dallas 1938). This was a venue case in which no act of any kind, much less a wrongful one, occurred in the county where suit was sought to be transferred. So, I suppose, the holding was dicta, or at least an alternative holding not necessary to a decision in the case. On what finger the current Texas Supreme Court may balance the issue is debatable. But if 402B imbodies a cause of action irrespective of fault for indirect representations, it is not a leap in logic to permit the imposition of liability for grossly negligent or fraudulent conduct.
SCIENCE
Conservative science uses statistics to establish relationships between environmental toxin and disease. Using a null hypothesis, the probability of no relationship may not exceed 5% in order for a non-random link to be established. For example, Henniken found that the probability of no relationship between silicone implants and a connective tissue disease was .00015 or .015%. The result was written O.R. = 1.24; C.I. = 1.08 - 1.41; P= .00015. What this means is that the groups being studied showed a 24% increase of disease in the implant group when compared to the non-implant, control group, and that if the study were done 100 times, at least 95% of the time the result would show an increase between 8% and 41%. The level of certainty that this result is not a random aberration is one and one half chance in 10,000. The authors of this study concede that it could be biased because it was a self reported study. There was to be a follow-up to determine whether the self-reporting by health care professionals was accurate, but I think I can say with a high degree of certainty that we will never see a follow-up study.
TEXAS SCIENCE
In an historic decision, Texas now requires an odds ratio greater that 2 and a certainty of at least 95% to establish a causative relationship between toxin and disease. Under this standard (the Havner case standard), it is unlikely than any toxin can cause disease. The rationale is said to be that the incidence of disease in the toxin exposed group must be 200% greater in order to establish that it is more likely than not that a particular person's disease was caused by the toxin. To any self-respecting statistical scientist this rationale is hog wash or worse. The court is saying that the Plaintiff must prove with virtual certainty that the toxin probably caused the disease. Beside mixing metaphors, the court overlooks that odds ratios, as such, are of no particular importance to an epidemiologist. It is the confidence interval which controls. Using the court's guidelines lets examine cigarette smoking and lung cancer. Epidemiologic studies in China show odds ratios which vary between 4.79 (squamous cell) and 1.02 (adeno) with confidence interials of 4.02 - 5.70 and .87 - 1.20, respectively.
Epidemiology has also examined the relationship between pancreatic cancer and smoking. The odds ratio for men is 1.6 with a confidence interval of 1.1 to 2.4. In women the odds ratio was 2.3 with a confidence interval of 1.4 to 3.5.
Also, let's examine, as did a Norwegian study, the relationship between lung cancer caused by manufactured cigarettes and such cancer caused by hand rolled cigarettes. That study showed a 90% increase (an odds ratio of 1.9), in lung cancer for hand rolled cigarette smokers compared to manufactured cigarette smokers. One last statistic before we roll these together. A smoker has a 33% chance of dying from cancers of any type.
Nevertheless, according to the most highly regarded text on the subject, cigarette smoking causes lung cancer, as well as laryngeal, oral, pharyngeal, esophageal, and bladder cancers. Cigarette smoking is an important contributing factor in the development of kidney and pancreatic cancer. It is associated with stomach and cervix cancer, and may be associated with leukemia, hepatoma, skin cancer (non-melanoma) and cancer of the anus and vulva. See Harrison's Principles of Internal Medicine, pp. 2517-2518 (14th ed.). Harrison's also reports that adeno-carcinoma is now the most common form of lung cancer. Id at p. 552.
Based on Havner, if I were to try an adeno-carcinoma case in Texas on behalf of a dead Chinese man, or a Japanese woman or a New Jersey woman, I'd never get to the jury. More importantly, because adeno-carcinoma represents about 1/3 of all lung cancers, what effect does that have on epidemiologic studies which don't differentiate cancer type? If we throw in the Norwegian study we can further confound the causation problem if the study doesn't differentiate between those that roll their own and those that don't.
Does the Texas court really mean to say that I could get to a jury in a smoke related pancreatic cancer case for a female but not for a male, and is the fact that a toxin is an "important contributing factor" insufficient to establish causation under Texas law? Lastly, does the mere 33% death rate for smoking induced cancer mean that it is not more probable than not that death was caused by cigarette related disease?
To a large measure, to state the argument is to refute it. Any medical epidemiologist will tell you that science requires 95% certainty before a valid risk-relationship may be said to exist because of the need for strong assurances before an agent is branded a toxin "generally." So, let's say a environmental substances causes a 30% increase in the incidence of brain cancer and the study has a 4% error rate. We know there are many causes of brain cancer including those deemed idiopathic. So what we know is that we have a 30% increase in the toxin exposed group when compared to a control group whose brain cancer component has ten (10) known other causes, plus the idiopathic. My first proposition is a simple one. If a scientist can eliminate the ten (10) known other causes which account for 50% of the brain cancers, he has established a greater than 50/50 chance that the substance caused the cancer in this instance. Second, even in the absence of elimination of other known causes, where a person is exposed to a known toxin and then develops the appropriate disorder, most practicing doctors will tell you that the simplest solution is probably the correct one. In law its known as K.I.S.S. (keep it simple stupid).
TEXAS LAW
What the court accomplished in Havner was seemingly a rejection of guidelines on causation laid down in modern cases, based upon "a sense of justice." Prior to 1952, Texas law required a plaintiff to prove the extent of his damages caused by each wrongdoer, even though this was not possible. This Robicheaux rule was eloquently dispatched by Justice Calvert in East Texas Salt Water Disposal, Co. v. Landers, 248 S.W. 2d 731, 734 (Tex. 1952):
"Thus at the time he filed his suit the Plaintiff in this case was confronted with a declared status of the law in this state which effectively relieved the two Defendants of the consequences of their wrongs and required the innocent Plaintiff to suffer his injuries without recompense. Whether he insisted on standing on his joint and several suit against the Defendant or bowed to the ruling of the trial court and proceeded against the Defendants separately, he could not prevail. In other words, our courts seem to have embraced the philosophy, inherent in this class of decisions, that it is better that the injured party lose all of his damages than that any of several wrongdoers should pay more of the damages than he individually and separately caused. If such has been the law, from the standpoint of justice it should not have been; if it is the law now, it will not be hereafter. The case of Sun Oil Co. v. Robicheaux is overruled."
It should be of more than just historical interest that Justice Calvert regarded Landers as his most praised opinion on an issue of justice. Calvert, Here Comes the Judge, p. 154 (1977).
Recently, our Supreme Court reaffirmed a commitment to this sense of justice in Amstadt v. U.S. Brass Corp., 919 S.W. 2d 644, 654 (Tex. 1996). There, U.S. Brass and VanGuard were manufacturers of polybutylene plumbing systems and the celcon fittings used to secure the pipes together. Frequently the products were mixed when installed. While less than apparent the proof did not demonstrate whether the damage or system failure was the result of product from VanGuard or U.S. Brass. (See the Court of Appeals opinion 864 S.W. 2d at 630-633). Justice Cornyn (now General), for a majority of the Court, resolved the issue as follows:
"The court of appeals held that for certain plaintiffs in Barrett, 'there is no evidence from which the jury could have allocated the liability as it did between U.S. Brass and Vanguard,' and that accordingly, 'there was no evidence of causation of damage to the homes and personal property' of those plaintiffs. 864 S.W.2d at 633. If, however, there was evidence that U.S. Brass' negligence was a proximate cause of the plaintiffs' damages, U.S. Brass' responsibility for that damage did not evaporate if the jury erred in apportioning liability between U.S. Brass and Vanguard. If the injuries arising from the plumbing system could not be apportioned with reasonable certainty, then the plaintiffs' injuries were indivisible, and the defendants are jointly and severally liable for the whole. See Landers v. East Tex. Salt Water Disposal Co., 151 Tex. 251, 248 S.W.2d 731, 734 (1952). Because the plaintiffs established the elements of their negligence claims, they are entitled to recover from U.S. Brass for its negligence. We accordingly reverse the court of appeals' take-nothing judgment as to the plaintiffs' negligence claims, and remand those claims to the trial court. At retrial, U.S. Brass will have the burden of apportioning its liability for the plaintiffs' injuries. If U.S. Brasscannot establish its percentage of liability, and thus remains liable for the whole, the trial court should credit U.S. Brass for the amounts the plaintiffs received in settlement from the other joint tort feasors. See Riley v. Industrial Fin. Serv. Co., 302 S.W.2d 652, 656 (Tex.1957)." 919 SW 2d at 654.
Shortly before the Amstadt decision our Supreme Court explained:
"The only exception we have recognized to our longstanding causation principles is where the inextricable combination of joint tort feasors combines to cause harm in a manner where individual responsibility can not be fixed. (Citing Landers). In these situations, liability is fastened only after a judicial determination that the claimant's injuries were caused by someone, but proof of that responsibility is impossible. Such cases are clearly distinguishable from those where the preponderance of the evidence suggests that no known person was at fault." Kramer v. Lewisville Memorial Hospital, 858 S.W. 2d 397, 405-406 (Tex. 1993).
But what about cases like Harris v. Cleveland, 294 S.W. 2d 235 (Tex. Civ. App. 1956 - Galveston, writ dismissed)? There four (4) welders worked on a job doing the same type of work. One of the welders failed to fill in an access hole, which led to the failure of a drilling venture. The plaintiff, however, could not show, at least according to his pleadings, which of the four (4) failed to fill in the hole. The Court of Appeals decided a venue issue adverse to plaintiff on the ground that plaintiff had not pleaded that the defendants were joint tort feasors and no proof was submitted that the local defendant was a proper party. On rehearing the court pointed out that it was not pre-empting the possibility of a jury at a trial determining which of the welders was at fault. Id. at 243. Compare Louisville Gas & Elect. V. Nall, 178 KY 33, 198 S.W. 745 (1917). In Nall the court let stand a verdict against one of two defendants for a single wrong where the evidence did not demonstrate which of the defendants was at fault.
In Riley v. Industrial Finance Service Co., 302 S.W. 2d 652 (Tex. 1957), the plaintiffs were husband and wife who had been harassed by 17 different moneylenders. These collection efforts caused an assortment of illnesses ranging from upset stomach to exacerbation of palsy. The collection efforts went on over a period of years, in varying intensity and effect on the plaintiffs. At trial against only one of the lenders plaintiffs testified that they really could not say which of the lenders caused which harm. The jury found damages proximately caused by the one lender on trial. But the jury also found that acts of "some" of the other lenders were a "proximate cause" of damage to the plaintiffs. The Supreme Court found this to be a Landers situation and remanded the case for another trial at which the damage issue should include the "total" damages suffered by plaintiffs. Id at 310-11. See also Marshall v. Joshe's, Inc., 581 S.W. 2d 192 (TCA - S.A. 1979, no writ).
What is it that concerned the Havner Court so much that a burden has been placed on the plaintiff which cannot be met? If the concern is that a culpable defendant might have to answer for damages possibly caused by something or someone else, Landers, of course, tells us that as between culpable defendants, the law is not so concerned that the plaintiff is denied recovery because of the uncertainty. What, please tell me, is different if a culpable defendant is made to answer for the harm caused by an innocent or not liable third party. Indeed this appears to be what was actually happening in Riley, supra, and Marshall v. Joske's, Inc., supra.
If all of this sounds like the imposition of liability for "increased risk" which in turn sounds like "loss of a chance" the solution is a simple one. Perhaps a legally liable party should be required to pay only for its own percentage of causation, if it has the burden of proof, and it meets that burden. Surely it is no more distasteful for a defendant to pay money which might be owed by a guilty party than it would be to pay for damages which might have been caused by an innocent participant.