Opinion
Argued April 10, 1990 —
Decided September 4, 1990.
Appeal from Superior Court, Law Division, Camden County
Before Judges ANTELL, ASHBEY and STEIN.
James C. Gavin argued the cause for appellant ( Gavin Gavin, attorneys; James C. Gavin on the brief).
Therese M. Keeley argued the cause for respondents Celotex Corporation and Owens-Illinois, Inc. ( McCarter English, attorneys; Therese M. Keeley and Rosanne C. Kemmet, on the brief).
Donald M. Kaplan argued the cause for respondent Owens-Corning Fiberglass Corporation ( Horn, Kaplan, Goldberg, Gorny Daniels, attorneys; Donald M. Kaplan and Raymond R. Chance, III, on the brief).
Plaintiff sues defendant manufacturers and distributors of asbestos products for her husband's death of colon cancer which plaintiff claims was caused by exposure to defendants' products. The suit, seeking compensatory and punitive damages, is grounded in theories of strict liability, negligence, breach of warranty and intentional tort.
Decedent had been employed as a maintenance man and pipe insulator at the Bayonne Terminal Warehouse from 1956 until December 1981. From 1956 until 1972 his work as an insulator consisted of removing, cutting, fitting and installing asbestos insulation. It is not denied that he developed cancer of the colon and, after surgical intervention, died December 10, 1982, at the age of 58.
Plaintiff now appeals from an order dismissing her complaint for lack of evidence establishing asbestos exposure as the cause of the disease. She contends that the trial court erroneously found that the expert testimony upon which plaintiff relied to prove causation amounted only to a factually groundless "net opinion." Plaintiff also asserts as error the trial court's refusal to permit a non-medical witness to testify that asbestos exposure was the cause of decedent's colon cancer, and in severing for trial plaintiff's negligence-based claim from the claim grounded in strict liability.
Asserted, too, during the trial was plaintiff's claim for compensatory and punitive damages for decedent's pleural thickening, also said to be asbestos-related. The trial court dismissed this claim for failure to show that the condition constituted an injury for which damages could be allowed, since it was unaccompanied by any diminution of function and decedent was not even aware of its existence. Plaintiff asserts that notwithstanding the lack of measurable harm, the claim for exemplary damages should have been submitted to the jury.
Defendants cross-appeal from a ruling of the trial court receiving into evidence a government publication commenting on the relationship between asbestos exposure and the development of colon cancer.
It is settled that while qualified expert testimony may be received to support a claim, it must have a foundation in the evidence and not be based on mere speculation or possibility. The term "net opinion" has been used to describe an expert opinion without a factual basis. "The `net opinion' rule appears to be a mere restatement of the established rule that an expert's bare conclusions, unsupported by factual evidence, is inadmissible." Buckelew v. Grossbard, 87 N.J. 512 , 524, 435 A.2d 1150 (1981); Vuocolo v. Diamond Shamrock Chem. Co., 240 N.J. Super. 289 , 299-300, 573 A.2d 196 (App.Div. 1990), certif. den. ___ N.J. ___ (July 17, 1990); Parker v. Goldstein, 78 N.J. Super. 472, 189 A.2d 441 (App.Div. 1963), certif. den. 40 N.J. 225, 191 A.2d 63 (1963). "This need for supporting data and a factual basis for the expert's opinion is especially important when the opinion is seeking to establish a cause and effect relationship." Rubanick v. Witco Chemical Corp., 242 N.J. Super. 36, 49, 576 A.2d 4 (App.Div. 1990). The principal question is whether the opinion evidence proffered by plaintiff herein satisfies the foregoing precepts.
The medical witness upon whom plaintiff relied for proof of causation was Dr. Joseph W. Sokolowski, Jr., a physician specializing in asbestos-related disease. Dr. Sokolowski never examined or treated decedent. From his study of decedent's personal, occupational and medical histories, which were taken from hospital records and plaintiff's answers to interrogatories, he testified to his belief that decedent had contracted colon cancer from exposure to asbestos. He based this on the decedent's "history of occupational exposure, the absence of other risk factors and the epidemiological evidence in the literature," adding "there is no other predisposing factor for the occurrence of carcinoma of the colon in an individual who is the age of Mr. Landrigan." Taken together with the epidemiological literature, the fact that asbestos is the only identifiable risk factor established it, in Dr. Sokolowski's mind, as the causative agent. Although he later acknowledged that the cause of most cases of colon cancer is unknown, he answered "No" when asked if decedent would "have gotten colon cancer" had he not been exposed to asbestos.
The witness explained that asbestos fibers enter the human system by inhalation into the upper airways. From there they may migrate to other organs or be regurgitated and swallowed into the gastrointestinal tract. While some of the fibers will be excreted in the feces, others are absorbed into the intestinal walls where, after a lengthy latency period, they may alter the cellular structure of the tissue to cause malignant tumors. In such cases fibers will be seen in the patient's feces and intestinal tissue. However, in this case no reference to such findings appears in the hospital records. Dr. Sokolowski explained that this was because the necessary special diagnostic equipment for such an investigation was never employed.
The doctor also testified that x-rays showed a "minimal degree of pleural thickening," that is, asbestos-caused scarring of the outer lining of the lung which resulted in no loss of pulmonary function. This discovery was made by Dr. Sokolowski during the trial. The decedent was not aware of the condition during his lifetime.
In considering whether plaintiff's circumstantial evidence suffices to prove causation it is relevant to note that colon cancer is the second most common form of cancer in American males. As Dr. Sokolowski testified, there are approximately 150,000 cases a year, of which about half result in mortalities. Substantial variations in its incidence occur according to geographical location, not only within the United States, but throughout the world. In this case, the decedent's malignancy took the form of an adenocarcinoma, the characteristic form of 99% of all cases of colon cancer; it was physically indistinguishable from almost all other such cases.
While Dr. Sokolowski agreed that the cause of most colon cancers in the general population is unknown or speculative, he attempted to qualify the point by adding "if you don't have an identifiable risk factor." The risk factors to which he apparently alluded are cigarette smoking, diet, beer drinking, certain predisposing diseases such as ulcerative colitis, familial polyposis, Crohn's disease and, of course, family history. We extract from this the witness's rationale that where a risk factor is present it is, ipso facto, the etiologic agent of the disease. Although not stated by Dr. Sokolowski in so many words, he seems to be saying that risk exposure equates with causation, a proposition which we find legally untenable.
Apart from the fact that he did not say at what degree of intensity mere "exposure" matures into a meaningful risk, Dr. Sokolowski made no allowance for the presence of other causes which have not yet been identified by medical science but which, as he acknowledged, account for most cases of colon cancer. With this must also be considered his own testimony that exposure to asbestos does not necessarily result in an asbestos-related disease. We therefore find no basis for the witness's postulate that the cause of a colon cancer is no longer unknown once a risk factor is identified. We conclude similarly as to his other postulate, "the absence of other risk factors." One cannot rule out the presence of other risk factors without knowing what those factors may be.
In defending his opinion, Dr. Sokolowski acknowledged that in determining whether asbestos fibers are operative in the etiology of a colon cancer the type and length of occupational exposure is important. He described this as the dose-response factor. As he put it, "you're more likely to see colon cancer in individuals who have the higher degree of exposures."
About decedent's occupational exposure, all Dr. Sokolowski knew was that he had worked as an insulator and maintenance employee at the Bayonne Terminal "and he was exposed to asbestos products." The doctor never saw decedent's job description, did not know how much of his time was spent working indoors and how much outdoors, how much time was spent working with asbestos "as opposed to doing regular maintenance," what kind of asbestos he was exposed to, what types of products, what percent of the products contained asbestos and what percent contained other materials. In short, he had no knowledge as to the degree of decedent's exposure. With these many significant gaps in the witness's working hypothesis, the trial court properly refused to permit the jury to credit his medical opinion that decedent's exposure to asbestos was causally linked to decedent's colon cancer and his unfortunate demise.
Major reliance for Dr. Sokolowski's opinion on causation was placed on epidemiological studies and articles which showed an association between colon cancer and exposure to asbestos. Epidemiology seeks to determine a relationship between a disease and a factor, such as a toxic substance, suspected of causing it. Brock v. Merrell Dow Pharmaceuticals, Inc., 874 F.2d 307, 311 (5th Cir. 1989), reh'g den. en banc, 884 F.2d 167 (5th Cir. 1989), cert. den. ___ U.S. ___, 110 S.Ct. 1511, 108 L.Ed.2d 646 (1990). The epidemiologist's investigation is carried on by comparing the disease experience of a population exposed to the factor with other populations not so exposed. Using statistical methods, the ultimate goal of an epidemiological study "is to draw a biological inference concerning the relationship of the factor to the disease's etiology and/or to its natural history." Black and Lilienfeld, Epidemiologic Proof in Toxic Tort Litigation, 52 Fordham L. Rev. 732, 750-751 (1984). It bears emphasis that such studies focus on the impact of suspected health related factors upon human populations. They do not attempt to determine the impact of these factors upon particular individuals. At least when performed by governmental agencies they are regarded as useful proofs. Ellis v. International Playtex, Inc., 745 F.2d 292, 300-301 (4th Cir. 1984); In re Agent Orange Product Liability Litigation, 611 F. Supp. 1223 , 1239-1240 (E.D.N.Y. 1985), aff'd 818 F.2d 187 (2nd Cir. 1987), cert. den. 487 U.S. 1234, 108 S.Ct. 2898, 101 L.Ed.2d 932 (1988). Their statistical import may be considered with other proofs in toxic tort litigation to help determined the suspected substance's effect on individuals. Mauro v. Raymark Industries, Inc., 116 N.J. 126 , 145, 561 A.2d 257 (1989); Thompson v. Merrell Dow Pharm., 229 N.J. Super. 230 , 249-250, 551 A.2d 177 (App.Div. 1988).
An epidemiological study compares the incidence rates among exposed and unexposed groups to see whether the rates differ. The term "expected mortality" refers to the rate of mortality for the general public, unexposed to the suspected risk. The term "observed mortality" refers to the rate of mortality in the study group. By dividing the expected mortality into the observed mortality the investigator arrives at a "risk-ratio" or "relative risk." From that is derived the attributable risk, "representing the proportion of the disease that is statistically attributable to the factor. * * * The attributable risk, therefore, is a composite measure that takes into account both the relative risk of disease if exposed and the proportion in the population so exposed." Black and Lilienfeld, supra, at 761. Relative risk shows the strength of the association between the suspected factor and the disease. Attributable risk is the percentage of decline of the disease incidence in the population which would result from elimination of the suspected factor. Ibid.
The process of arriving at the relative risk and the attributable risk was described in Cook v. United States, 545 F. Supp. 306 (N.D.Cal. 1982). There, plaintiffs sought damages for having suffered from Guillain-Barre Syndrome (GBS) which they claimed was caused by their inoculation with a federally sponsored swine-flu vaccine. The court held that the plaintiffs' epidemiological evidence was legally insufficient to prove causation because it failed to show a relative risk in excess of twice the upper limit of the baseline risk. The court explained the requirement of a relative risk in excess of two as follows:
Whenever the relative risk to vaccinated persons is greater than two times the risk to unvaccinated persons, there is a greater than 50% chance that a given GBS case among vaccinees of that latency period is attributable to vaccination, thus sustaining plaintiff's burden of proof on causation. [ Cook, 545 F. Supp. at 308 ].
It then presented the following illustration of how the process works:
This fact is readily illustrated by a hypothetical example. Suppose the relative risk for vaccines nine weeks after vaccination is two — i.e., that they are twice as likely to experience onset of GBS after that interval as are persons in the unvaccinated population during the calendar week. If fifty GBS cases occur among a million unvaccinated persons that week, then a hundred cases would be expected among a million nine-week vaccinees. Of that hundred, fifty would have been expected without vaccination, while the other fifty are explained only by the event of vaccination. Thus, the likelihood that a given nine-week vaccinated case of GBS is attributable to vaccination is 50%.
Similarly, if the relative risk of GBS to nine-week vaccinees is four, then 75% of all nine-week vaccinees are vaccine-linked. Once the relative risk rises above two, it becomes more probable than not that a given case was caused by the vaccine. [ Id. at 308 fn. 1].
Other courts which have considered the question also recognize the requirement that an epidemiological study show a relative risk in excess of two in order to be statistically significant. See Manko v. United States, 636 F. Supp. 1419 , 1434 (W.D.Mo. 1986) aff'd and remanded, 830 F.2d 831 (8th Cir. 1987) (relative risk of two, or less, means exposure is not the probable cause of disease claimed); Marder v. G.D. Searle Co., 630 F. Supp. 1087 (D.Md. 1986), aff'd other grounds sub nom. Wheelahan v. G.D. Searle Co., 814 F.2d 655 (4th Cir. 1987) (risk ratios of 1.3 and 1.9 held insufficient to allow for a finding of probable causation of harm by defendant's intra-uterine device); In re Agent Orange Product Liability Litigation, 597 F. Supp. 740 , 785 (E.D.N.Y. 1984), aff'd 818 F.2d 145 (2d Cir. 1987) (plaintiffs must prove at least a twofold increase in evidence of disease allegedly caused by exposure to suspected toxic substance).
The epidemiological study chiefly relied on by Dr. Sokolowski was set forth in "the significant article" by Dr. Irving Selikoff which appeared in the Annals of the New York Academy of Sciences in 1979. It reviewed the results of a study of the occupational history and exposures of 17,800 insulators located in the United States, principally in the northeast, and in Canada. The study disclosed 59 deaths from colon cancer in the study population of which 38 were "expected." The relative risk of the study group was found to be 1.55, reflecting an attributable risk of approximately 35%. Under standards accepted in the well-reasoned decisions cited above, the strength of the association shown by such a risk ratio does not sufficiently demonstrate causation so as to validate the witness's opinion.
We have also considered the cases cited in Mauro v. Raymark Industries, Inc., 116 N.J. at 139-141, 561 A.2d 257.
Dr. Sokolowski's allusions to other epidemiological studies of asbestos exposure and colon cancer which he examined in preparation to testify were for the most part conclusory in form. Some studies were positive, some negative. As he acknowledged, there were "many studies that show no statistically significant increase in colon cancer in workers exposed to asbestos." The closest Dr. Sokolowski came to explaining the studies and why they differ in result was when he tried to rebut the inference, otherwise to be drawn, that the import of the negative studies is that asbestos "doesn't cause cancer of the colon." He then briefly stated that "you have to consider the degree of exposure of these various groups," the kind of work being performed by the members of the study group and the kind of asbestos to which they had been exposed. He added that "many of the articles really don't consider the latent period." Nothing more specific was said. The witness made no reasoned statement as to why he chose to accept the positive results and reject the negative. The record is silent about the methodology employed in the studies to control the variables, such as other risk factors and predisposing elements, so as to minimize distortion in the final analysis. Nothing was said about the type and extent of exposure of the individuals in the study groups or, as we said before, the type and extent of decedent's exposure. Therefore, comparisons were impossible. Such omissions in the record left the studies, which were so superficially discussed, valueless as a factual predicate to Dr. Sokolowski's opinion for causation. Washington v. Armstrong World Industries, Inc., 839 F.2d 1121 (5th Cir. 1988), is instructive. There too plaintiff's deceased husband died from colon cancer allegedly caused by asbestos exposure. The trial court granted summary judgment on the basis of affidavits from three physicians, one of whom asserted that there was "no pathologic basis to even speculate that asbestos was a factor in the development of this tumor" absent a demonstration of asbestos or fibers in the tissue. Id. at 1122. Plaintiff moved for reconsideration on the basis of an affidavit executed by her expert, Dr. Comstock, who concluded "that there was a reasonable medical probability that asbestos exposure caused Mr. Washington's cancer" based on the facts that decedent "had been exposed to asbestos for 32 years and that there was a statistically significant association between asbestos dust inhalation and cancer of the colon." Ibid. As to the absence of asbestos fibers in samples of the decedent's tissue, the doctor stated that the "customary examination procedures used in pathology may not have been sensitive enough to detect asbestos fibers." Ibid. This, plaintiff contended, sufficed to create a fact question as to causation.
The witness's broadly stated conclusions do not allow for the kind of factual analysis needed to determine whether the studies reliably show the probability of a link between colon cancer and asbestos exposure. For one example of the critique required of a fact finder in order to extract meaning from the statistical proofs see Cook v. United States, supra.
The trial court concluded that Dr. Comstock's affidavit was pure speculation and fundamentally unreliable and declined to disturb the summary judgment already entered. On review, the Circuit Court of Appeals affirmed. It pointed out that the Comstock affidavit amounted to nothing more than a statement that a link between colon cancer and asbestos exposure is "statistically probable," but did not rise to the level "of proving that the deceased sustained an actionable injury caused by the acts or omissions of the defendants." (emphasis in original) Id. at 1123. Thus, the court ruled that although Comstock's "data linking asbestos exposure to colon cancer must be accepted as accurate for the purposes of the motion, the affidavit failed to demonstrate a link between this data and Washington's cancer." 839 F.2d at 1124.
It has been wisely stated in toxic tort litigation that courts must be particularly careful "when assessing the sufficiency of causation evidence, where causation evidence is wholly circumstantial." Marder v. G.D. Searle Co., 630 F. Supp. at 1089 (citing Lovelace v. Sherwin-Williams, 681 F.2d 230, 242 (4th Cir. 1982); Lohrmann v. Pittsburgh Corning Corp., et al., 782 F.2d 1156 (4th Cir. 1986)). Although expert witnesses are frequently indispensable in such cases, this does not mean that a court "must accept uncritically any sort of opinion espoused by an expert merely because his credentials render him qualified to testify." Richardson v. Richardson-Merrell, Inc., 857 F.2d 823, 829 (D.C. Cir. 1988), cert. den. ___ U.S. ___, 110 S.Ct. 218, 107 L.Ed.2d 171 (1989). "[C]ourts must critically evaluate the reasoning process by which the experts connect data to their conclusions in order for courts to consistently and rationally resolve the disputes before them." Brock v. Merrell Dow Pharmaceuticals, Inc., 874 F.2d at 310.
Applying these precepts, we conclude that the record presented for review provided no factually supported basis upon which the jury might have found a causal relationship between the colon cancer which took decedent's life and his exposure to asbestos fibers.
We appreciate that plaintiff should not be held to the standards of scientific certainty in her proofs. But it appears to us that the answer to the question presented under the circumstances of this case lies beyond the outskirts of medical knowledge, and plaintiff's proofs are simply incapable of engaging the reasoning processes of a jury. Decedent's occupational exposure to asbestos fibers during his life, while a suspicious circumstance, does not supply its own incriminating link to decedent's last illness, and the fact that causation cannot be reasonably demonstrated by presently available evidence is no justification for allowing a jury to guess its way to such a result. To decide otherwise would disserve our obligation to see that responsible judgments are reached in these difficult cases on the basis of intelligible standards, and not caprice.
Although we agree that the trial court erred in severing plaintiff's strict liability and negligence claims for trial, our disposition of this appeal would not be altered by the two claims having been tried together. The evidence of proximate cause would have been the same.
During the trial, plaintiff offered Dr. Joseph Wagoner as an expert witness to testify as to the cause of the decedent's colon cancer. The trial court rejected the offer on the ground that Dr. Wagoner was not qualified by medical training to determine the fact in question. We agree with the ruling under review. Dr. Wagoner's qualifications as an epidemiologist and biostatistician did not endow his opinion as to proximate cause with the expertise necessary to "assist the trier of fact to understand the evidence or determine [the] fact in issue." See Evid.R. 56(2). As we noted earlier, epidemiology deals with the movement of different diseases within human populations. It does not address questions of specific causation in the individual case. While epidemiological information, taken together with other medical facts, may be useful to a physician in forming a particular diagnosis or in determining the etiology of an illness, court determinations as to such matters cannot be based on an expert opinion which rests on the application of statistical skills and studies alone.
Dr. Wagoner's expertise was in the statistical study of the incidence of disease over large population groups. He was qualified to testify as to the relationship between asbestos exposure and the development of colon cancer, but not as to whether asbestos fiber was the causative agent of decedent's illness. Cases cited by plaintiff do not support a contrary result.
In Loudermill v. Dow Chemical Co., 863 F.2d 566 (8th Cir. 1988), the witness was a toxicologist with 12 years experience in determining the cause of death at autopsies and was a consulting staff member for emergencies at a hospital. Thus, he possessed the requisite skill in determining the cause of death in particular patients, and his testimony was properly admitted.
Rubanick v. Witco Chemical Corp., supra, was decided on the basis of an in limine hearing in which the Law Division had ruled that plaintiff's proposed expert witness would not be permitted to testify that decedent's colon cancer had been caused by defendants' toxic contaminant. Rubanick, 242 N.J. Super. at 40, 576 A.2d 4. In reversing, the Appellate Division noted that the witness held a Ph.D. in biochemistry and had 37 years of cancer research experience with the Sloan-Kettering Institute. At that time he was head of a research group primarily concerned with investigating the cause, treatment and diagnosis of colon cancer, and had published extensively on the topic of carcinogenesis. Id. at 43, 576 A.2d 4. In our view, his background in the physical sciences was materially different from the experience and training of Dr. Wagoner whose interest in the subject of carcinogenesis is that of a statistician.
In Gideon v. Johns-Manville Sales Corp., 761 F.2d 1129 (5th Cir. 1985), Dr. Wagoner, the same witness as in this case, was, as plaintiff states in her brief, allowed to testify to the "ability" of asbestos to cause cancer in a particular individual. As that court pointed out, Dr. Wagoner did not provide medical testimony and remained within his area of expertise. "He testified only on matters of epidemiology: the risk of cancer and reduced life expectancy associated with asbestosis." Id. at 1136.
The trial court dismissed plaintiff's punitive damage claim for decedent's pleural thickening because the condition did not constitute a disability for which compensatory damages could be allowed. The "minimal pleural thickening" which Dr. Sokolowski discovered during the trial by examining an x-ray of decedent's lungs consisted of benign scarring to the outer lining of the lung. It did not interfere with decedent's breathing nor in any other way impair his function. There was no involvement of the parenchyma, the substance of the lung, the thickening caused no discomfort and, since decedent was never made aware of its existence, not even an emotional distress claim was viable. Compare Mauro v. Raymark Industries, Inc., 116 N.J. at 137, 561 A.2d 257. We find no error in the court's ruling. Although punitive damages may be assessed "for an intentional tort involving egregious conduct whether or not compensatory damages are awarded," plaintiff must "at least" show "some injury, loss, or detriment." Nappe v. Anschelewitz, Barr, Ansell Bonello, 97 N.J. 37 , 51, 477 A.2d 1224 (1984).
On their cross-appeal defendants complain that the trial court erred in taking judicial notice of two government agency publications relating to the link between colorectal cancer and asbestos exposure. In light of our disposition of the appeal, we find it unnecessary to pass upon this issue.
Affirmed on the appeal. The cross-appeal is dismissed.