FINDINGS OF FACT AND CONCLUSIONS OF LAW
SUTTLE, Senior District Judge.
This action was brought pursuant to the Federal Tort Claims Act, 28 U.S.C. § 2671
et seq.,
and the National Influenza Immunization Program of 1976, 42 U.S.C. § 247b. The plaintiffs, John H. Padgett and Suzanne Padgett, seek compensatory damages for injuries alleged to have resulted from Mr. Padgett’s inoculation with the swine influenza vaccine. The parties have stipulated that John Padgett developed GuillainBarre Syndrome (GBS) on March 10, 1977, 116 days, or 16 weeks, after his inoculation with the swine influenza vaccine on November 14, 1976. The dispositive issue for this Court is causation.
This action was filed on February 7,1979, in the United States District Court for the District of Columbia for coordinated pretrial proceedings in
In Re Swine Flu Immunization Products Liability Litigation,
M.D.L. No. 330, Misc. No. 78-0040 (D.D.C.1979). On February 29,1980, the parties stipulated to transfer of this action to this court for local discovery and for trial. A non-jury trial on the issues of liability and damages was held from June 14 to 18 and August 30 to 31, 1982. The court now enters its findings of fact and conclusions of law in the form of this written memorandum.
I. PLAINTIFF’S MEDICAL HISTORY
Mr. Padgett was 30 years of age at the time of his inoculation with the monovalent swine influenza vaccine on November 14, 1976, at the local high school in Seguin, Texas.
On March 4, 1977, the plaintiff called his family physician, LeRoy Weiss, M.D., and reported to Dr. Weiss the symptoms of a cold or the flu. Dr. Weiss advised him to come to his office for a bicillin injection, which he did. On March 7, 1977, the plain
tiff called Dr. Weiss and informed him that the first injection had not relieved his symptoms. Dr. Weiss recommended that he return for another bicillin injection, which was administered by Dr. Weiss’ nurse.
Dr. Weiss’ medical records for the plaintiff’s March 4 and 7,1977, visits contain Dr. Weiss’ impression that the plaintiff’s condition was an upper-respiratory infection (URI).
On March 8, 1977, the plaintiff visited a general practitioner in Seguin, Texas, Joseph B. Gastring, M.D., and gave a history of “fever on and off for eight days, from 100 to 101 in the p.m., malaise, medication and fever.” Dr. Gastring also noted that a swine influenza vaccination had been administered to the plaintiff on November 14, 1976. His impression was “viral and/or nonpyogenic sinusitis.”
The medical records of Dr. Gastring are closest to the date of the URI, and the only medical records in which Mr. Padgett himself estimated the onset of his URI. He estimated that his fever had been of eight days’ duration. This court specifically finds that the symptoms of the plaintiff’s URI began on March 1, 1977.
On March 10, 1977, Mr. Padgett experienced the onset of neurological symptoms of GBS. He awoke with numbness of the face and tingling of his tongue and hands. The plaintiff went to Dr. Gastring’s office and complained of generalized numbness and difficulty breathing. He had difficulty with balance and areflexia in his lower extremities. Dr. Gastring told the plaintiff that he probably had Guillain-Barre Syndrome, a form of temporary paralysis, but he did not have any opinion as to the severity of it. Dr. Gastring took chest x-rays, told the plaintiff to continue taking antibiotics, take off from work, go home, rest, and call if anything major developed.
The plaintiff returned home and awoke the next morning, March 11, 1977, with difficulty getting out of bed, difficulty walking, difficulty breathing, and extreme weakness. His wife called Dr. Gastnng, who referred her to Walter F. Buell, M.D., a neurologist in. San Antonio. Dr. Buell admitted plaintiff to Santa Rosa Hospital.
The plaintiff’s condition was diagnosed as Guillain-Barre Syndrome shortly after admission. He was hospitalized with GBS at the Santa Rosa Hospital for five months from March 11, 1977, to August 11, 1977. He became totally paralyzed during the first stages of the disease and could breathe only with the assistance of a respirator. A tracheostomy had to be performed.
The paralysis became so complete that Mr. Padgett could not control his urine or hands or even blink his eyes. Once he lost control of his eyelids, he lost his sole source of communication, as he had worked out a system of blinks as signals. After 13 weeks he was taken off the respirator and put in intensive care. He had to endure extensive rehabilitation, which was extremely painful.
The plaintiff eventually returned to work on a limited basis in October of 1977. In March of 1978 he went back to work on a full-time basis but his job was changed as he was considered a risk to drive. The plaintiff had to have an operation after the five months in Santa Rosa Medical Center to remove a bladder stone caused by the catheterization. His total medical bills are more than $43,000.00. He still has weakness in his feet and foot drop.
II. GUILLAIN-BARRE SYNDROME
Guillain-Barre Syndrome is an acutely evolving, ascending paralytic disease of the peripheral nervous system
of unestablished etiology. Although the precise mechanism of GBS is unknown, it is generally thought that the disorder represents an aberrant immune response.
In GBS patients, lymphocytes attack the myelin, the fatty substance that surrounds or “insulates” the peripheral nerves. The resulting “shorting out,” or demyelination, of the nerves impairs their ability to con
duct electrical impulses from the brain that control the reflexes and movement of certain muscles.
Attempts to isolate a viral or microbial agent responsible for GBS have failed. A respiratory, gastrointestinal infection or other infectious illness precedes the onset of symptoms of GBS within three weeks in approximately 60 percent of GBS patients. In many cases, no preceding event can be identified.
III. EXPERT TESTIMONY ON CAUSATION
The plaintiffs called Martin Lewis, M.D., a pathologist and Chairman of the Department of Pathology at Loyola University in Chicago; Donald V.T. Bear Ph.D., Professor of Economics at the University of California at San Diego; and Joseph B. Gastring, M.D., a general practitioner in Seguin, Texas.
The government called Thomas M. Mack, M.D., an epidemiologist who is Professor of Epidemiology in the Department of Pathology and Family and Preventive Medicine at the University of Southern California, Los Angeles, California; Edward A. Neuwelt, M.D., a neuroimmunologist and former Assistant Professor of Neurosurgery and Biochemistry at the University of Texas Medical Center at Dallas, currently in clinical and research practice in Portland, Oregon; W. Alan Hauser, M.D., a neuroepidemiologist who is Associate Professor of Neurology at the Gertrude A. Sergievsky College of Physicians at Columbia University in New York; Walter F. Buell, M.D., the plaintiffs treating neurologist and Associate Professor of Neurology at University of Texas Medical School in San Antonio, Texas; and LeRoy Weiss, M.D., the plaintiff’s family physician and a general practitioner in Seguin, Texas. The government also called, by videotape deposition, several physicians who testified in the multidistrict proceedings.
IV. PLAINTIFFS’ EVIDENCE ON MEDICAL CAUSATION
The plaintiffs offered three pieces of evidence in an effort to show that Mr. Padgett’s GBS was caused by his swine influenza vaccination 16 weeks earlier: (1) the videotape deposition testimony of Dr. Lewis; (2) the deposition testimony of a general practitioner, Dr. Gastring; and (3) the testimony of Donald V.T. Bear, Ph.D.
A. PLAINTIFFS’ EXPERT, MARTIN LEWIS, M.D.
Martin Lewis, M.D., testified for the plaintiffs on causation. Dr. Lewis is a pathologist. He is not a neurologist or an epidemiologist. A review of his curriculum vitae reveals that nearly all of his publications are on cancer.
Although his testimony centered on neuroimmunology, he has never authored a single article on GBS, peripheral neuropathy (which includes GBS), or autoimmune reactions and GBS. Nor has he written on neurological disorders following vaccination. Quite simply, Lewis has not published on the subject of neuroimmunology.
Dr. Lewis based his belief that the swine flu shot caused the plaintiff’s GBS on two factors. First, he testified that the swine flu shot primed or sensitized the plaintiff’s immune system and that the URI the plaintiff developed in March of 1977 was a secondary response building upon the previous response and resulted in a rapid onset of the GBS. (Lewis 29:14-23) Second, Dr. Lewis testified that it was inconceivable from what he knows about GBS that an infection 6 days before the onset of the GBS could be the prime cause of the GBS. (Lewis 30:12-16)
1.
Dr. Lewis’ Sensitization Theory Is Speculative
The basis for Dr. Lewis’ sensitization theory is that the priming factor (the swine flu vaccine) and the secondary factor (the URI) contain the same or similar proteins. (Lewis 168:14-20) However, he admitted that he did not know what proteins were in the swine flu vaccine:
I have no idea what the protein was that gave rise to this, nor does anyone else for that matter ...
Lewis 169:25-170:1.
See also
168:14-169:5. Dr. Lewis assumed that the proteins in the vaccine and the URI were the same because “[i]t’s the best assumption I can make.” (Lewis 170:1-4)
Dr. Lewis also testified that his proposed primary and secondary immune responses were caused by the same or similar virus. (Lewis 34:20-25) However, he did not identify the virus in the vaccine or the URI, or give any basis for his assumption that the two were similar or identical.
The immunologist who testified for the Plaintiffs’ Steering Committee during the MDL proceedings, Joseph Bellanti, M.D., characterized the plaintiffs’ sensitization theory as “speculative.” (Bellanti 458:15-459:1) Dr. Hauser, the defendant’s expert neuroepidemiologist, testified that there is no support in the medical literature for the sensitization theory, and that the Michigan epidemiologic study refutes it. (Hauser 81:9-82:6).
2.
Dr. Lewis’ Testimony That the Plaintiff’s URI Occurred Too Near the Onset of GBS to Have Been the Cause Is Directly Contrary to the Medical Literature
The other basis of Dr. Lewis’ opinion was that it was inconceivable that a URI with onset six days before GBS could have been the prime or sole cause of the GBS. (Lewis 30:12-16) He at first testified that, for the URI to have precipitated the plaintiff’s GBS, it should have occurred 2 to 3 weeks (14-21 days) before the onset of the plaintiff’s GBS. (Lewis 143:22-144:6) When confronted with his testimony in
Alvarez v. United States,
495 F.Supp. 1188 (D.Colo. 1980), that he would expect to see the infectious illness occur 10 to 14 days before the onset of the GBS for it to be the only cause of the GBS (Lewis 145:25-146:22), Lewis agreed that 10 days was a reasonable period of time for the URI to cause the GBS. (Lewis 145:3-8; 162:17-21)
Dr. Lewis had assumed, based on a conversation with the plaintiff the day before his videotape trial deposition was taken, that the URI in this case had onset 6 days before the GBS — on March 4, 1977 (Lewis 156:7-13), rather than March 1,1977, as this court has found.
As a last resort, Dr. Lewis testified that patients usually recover from the infectious illness before the onset of GBS. (Lewis 145:3-7) When asked for support for his contention that the URI or other infectious illness must have cleared before the onset of the GBS for the URI to have caused the GBS, he could point only to Dr. Arnason’s Chapter 56 in Dyck, et al.
Peripheral Neuropathy
at p. 1110-2 (U.S. Exhibit 6; Lewis 162:22423):
Approximately half the patients give a history of an antecedent acute infectious illness that has usually cleared by the time neuropathic symptoms begin. Most commonly the antecedent illness is described as influenza-like with fever; sometimes patients mention an acute dysenteric episode. Less commonly an earlier episode of low back pain is recalled.
The interval between the prodromal infectious episode and the onset of neuropathic symptoms is variable; most frequently it is one to three weeks. On occasion the interval between infection
and neuropathy may be less than one week. In most instances the agent responsible for the prodromal illness remains unidentified. (Citations omitted)
In summary, Dr. Lewis’ sensitization theory is speculative, and his contention that the URI occurred too close to the onset of the plaintiff’s GBS is not supported by and is contrary to the medical literature.
B. PLAINTIFFS’ EXPERT, DR. GASTRING, M.D.
The plaintiffs read into the record the deposition testimony of Mr. Padgett’s general practitioner, Dr. Gastring. Dr. Gastring is not a neurologist. Mr. Padgett’s was only the second case of GBS he had seen in his entire career. Dr. Gastring referred the plaintiff to Dr. Buell, a neurologist, for the care and treatment of his GBS.
On direct examination Dr. Gastring opined that the plaintiff’s GBS was caused by the swine flu vaccine. On cross-examination, he demonstrated lack of familiarity with the precipitating events of GBS and the time interval between the precipitating events and the onset of GBS. (Gastring 349:20-350:16) Dr. Gastring’s lack of knowledge of GBS was further demonstrated by his testimony that GBS is a communicable disorder (Gastring 343:23-344:6), which it is not.
The most telling evidence of Dr. Gastring’s lack of familiarity with GBS was the fact that, when he diagnosed it in Mr. Padgett on March 10, 1977, he sent him home, even though Mr. Padgett was having difficulty breathing. It is most inadvisable to send home a GBS patient who is having difficulty breathing. A physician with knowledge of GBS would have placed the plaintiff in intensive care that day because of his respiratory difficulty. (Hauser 42:5-43:1)
After Dr. Gastring’s knowledge of GBS was tested on cross-examination and it became fairly obvious that he did not know very much about GBS, he testified that he had no opinion on whether plaintiff’s GBS was caused by the swine flu vaccine. (Gastring 346:8-10 and 357:10-15) Accordingly, the court finds that Dr. Gastring’s testimony is of no aid in establishing causation.
V. DEFENDANT’S TESTIMONY ON CAUSATION
Walter F. Buell, M.D., the plaintiff’s treating neurologist, testified that, based on his education, clinical experience, and knowledge of GBS, the swine influenza vaccination did not cause the plaintiff’s GBS. (Buell 290:12-19; 314:11-15; 315:14^-18) He stated that the plaintiff’s URI appeared to be the precipitating event. (Buell 297:19-23)
W. Alan Hauser, M.D., a neuroepidemiologist from Columbia University, testified that the swine influenza vaccination did not cause or contribute to the plaintiff’s GBS. (Hauser 78:4 — 14) He stated that Dr. Lewis’ testimony that the plaintiff’s URI occurred too close to the time of onset to have been solely responsible for the GBS was not supported by the medical literature. (Hauser 80:18-81:3) He knew of no medical evidence to support Dr. Lewis’ sensitization theory. (Hauser 81:8-82:6)
Edward A. Neuwelt, M.D., is a neuroimmunologist and author of the text Neuwelt, Clark,
Clinical Aspects of Neuroimmunology
(1978). He testified that, because the swine influenza vaccine was a killed vaccine as opposed to a live vaccine, there were no live particles to sustain an infection for more than a period of a few days. (Neuwelt 69:20-70:4) Therefore, there was no probability that the antigens that Dr. Lewis alleges sensitized the immune system would remain in the body for more than a few days. (Neuwelt 70:5-7) . Dr. Neuwelt knew of no experimental or theoretical basis to support the sensitization theory. (Neuwelt 91:15-94:19)
Dr. Neuwelt testified that the most common precipitating events in GBS are: URI’s, GI’s, common colds, influenza, and other viral and bacterial infections. (Neuwelt 73:8-25) He testified that GBS may have onset as few as three days after the preceding infectious illness, although it usu
ally takes four to seven days to occur (Neuwelt 76:3-6; 77:5-14), and can occur up to two months prior to the onset of the GBS. (Neuwelt 76:6-14) Dr. Neuwelt rejected Dr. Lewis’ notion that the URI occurred too close to the onset of the GBS for it to have been the sole cause, explaining that it is inconsistent with the medical literature on GBS and standard immunologic data that has been known for fifty years. (Neuwelt 77:15-24; 94:20-95:24) He also testified that it made no difference whether the plaintiff’s URI had cleared when he developed GBS. (Neuwelt 90:24-91:11)
Dr. Neuwelt testified that the swine influenza vaccination did not cause or contribute to the plaintiff’s GBS and that the sole precipitating event was the plaintiff’s URI. (Neuwelt 88:2-8) He characterized the scientific evidence in support of his opinion as “overwhelming.” (Neuwelt 88:9-11) As a neuroimmunologist, he knew of no clinical or experimental data that would support a relationship of a killed antigen inciting an immune response against one’s own body 16 to 17 weeks later. (Neuwelt 88:12-89:15)
Besides these live witnesses, the defendant put on several other expert witnesses by multidistrict videotape deposition. Each of these eminent experts gave his opinion as to the period of increased likelihood of contracting GBS following a swine influenza vaccination.
Peter J. Dyck, M.D., Professor of Neurology at the Mayo Medical School, estimated six weeks. (Dyck 37:19-25) Barry G.W. Arnason, M.D., Chairman of the Department of Neurology at the University of Chicago, estimated nine weeks at the outside. (Arnason 51-52) Alexander Langmiur, past Professor of Epidemiology at Harvard Medical School, estimated ten weeks. Neal Nathanson, M.D., Chairman of the Department of Microbiology at the University of Pennsylvania Medical School and formerly editor of the American Journal of Epidemiology, estimated ten weeks. Paul F. Wehrle, M.D., Hastings Professor of Pediatrics at the University of Southern California specializing in immunization and epidemiology, estimated 60 days.
VI. ULTIMATE FINDING ON MEDICAL ASPECTS OF CAUSATION
The court specifically finds the expert testimony presented by the defendant more persuasive on the medical aspects of causation in this case. The court rejects Dr. Lewis’ sensitization theory as speculative.
VII. EPIDEMIOLOGY
The National Influenza Immunization Program began on October 1, 1976. Before the vaccine administration began, a nationwide surveillance system was established to rapidly identify illness temporally related to vaccination. By December 2, 1976, when over 35 million doses of vaccine had been administered, two clusters of GBS in recent inoculees had been reported to the Center for Disease Control (CDC) in two states: Minnesota (four cases) and Alabama (three cases).
Following identification of the two clusters, CDC initiated an epidemiologic investigation to evaluate the possibility of a relationship between the GBS and the swine flu vaccine. By December 15, 1976, data from four states suggested that the incidence of GBS in vaccinees was approximately seven times greater than the incidence in non-vaccinees. This led to the termination of the National Influenza Immunization Program on December 16, 1976, and an expansion of the surveillance program for GBS. The results of the surveillance program showed an increased risk of GBS up to ten weeks after inoculation; however, the risk was not statistically significant after eight weeks. Schoenberger, et al.,
Guillain-Barre Syndrome Following Vaccination in the National Influenza Immunization Program, United States, 1976-1977,
A.J. Epi. 160:105, 1979.
Epidemiology is the study of the patterns and causes of disease in human populations. Epidemiologists investigate factors or events that are associated with disease in an effort to determine which of the factors or events are causing the disease.
Epidemiologists are the professionals who are trained to study the cause and effect relationship. They are trained to work with groups of people and populations, analyze reports of disease in relation to the populations, determine incidence rates in the groups, and interpret the differences in various incidence rates. Epidemiologists are qualified to evaluate the level of association, and explanations for the association other than causality.
There are three components in evaluating causation of disease: (1) the level of the association between an event and the disease; (2) the biologic credibility of the purported association; and (3) whether there is an alternative explanation for the purported causality that has more credibility than that of causation. In evaluating the first component — the strength of association— chance must be excluded. If the association is not explained by chance alone, then it may be explained by a causal relationship or some other relationship. In evaluating the second component of causality — biologic credibility — the purported association is compared with medical knowledge on the subject. In evaluating the third component of causality — whether there is an alternative explanation for the purported causality — epidemiologists must exclude three alternative explanations: chance, confounding, and bias. Chance has been previously explained. Confounding, not applicable here, would explain a statistically significant relationship between lung cancer and yellow fingers.
Bias is an error in the actually observed rate of a condition that comes about by virtue of the means of counting or the means of establishing the criteria; i.e., an irregularity in the surveillance mechanism. Bias may come from the people who have the disease, the physicians diagnosing or reporting it, or the persons performing the surveillance. For example, if a radio station announces that it wishes to count the number of cases of cancer in a community in relation to a chemical dump site, epidemiologists expect that more people who live near the dump site will respond to the call than people who live far away from the dump site. The conclusion could be drawn by the people doing the study that there is a relationship between living near the dump site and developing cancer when in fact the association resulted from increased motivation to report and not causation.
The strength of the association between GBS and the swine influenza vaccination is very high — on the order of a seven-fold increase for all periods combined, according to the Schoenberger study. That there is a tremendous increase in the relative risk concentrated primarily within the first five weeks after vaccination implies causality. But the fact that a temporal association exists between vaccination and GBS also suggests that at some point the temporal association becomes so attenuated that causation cannot be attributed to the vaccine.
The level of association between an event — such as vaccination — and a disease — such as GBS — is determined by comparing risks in comparable populations.
Assuming 100% reporting of all GBS cases,
the relative risk to a vaccinee as opposed to a non-vaccinee is determined by dividing the attack rate ámong vaccinees in a given population by the attack rate among nonvaccinees in that population. For example, if the attack rate among vaccinees three weeks after inoculation were 4 per million and if the attack rate among non-vaccinees were 1 per million, the relative risk to a vaccinee three weeks after inoculation would be 4. Of course, this computation assumes that there is a 100% reporting of GBS cases in both the vaccinated and unvaccinated segments of the population that we are studying.
From the relative risk, we can calculate the probability that a given case of GBS was caused by vaccination. The probability is equal to the relative risk minus one, divided by the relative risk. In the example above, then, the probability that vaccination caused GBS in a three-week vaccinee would be (4_1)/4, or %. In other words, in a large population of three-week vaccinees, 25% of the GBS cases would have occurred in the absence of vaccination and 75%
would not have occurred but for the vaccination. A relative risk of 2 or greater, then, means that the probability that vaccination caused a particular case of GBS is better than 50%. Hence, a relative risk of 2 or greater would indicate that it was more likely than not that vaccination caused a case of GBS.
To obtain an ideal calculation of the relative risk of vaccination, however, we need 100% reporting of all GBS cases in both vaccinees and non-vaccinees. In an imperfect world, such ideal reporting is nonexistent. To correct for these imperfections, the plaintiffs’ expert, Dr. Bear, made certain assumptions. The court finds that these assumptions are epidemiologically unsound and also contrary to common sense.
A. PLAINTIFF’S EXPERT, DONALD V.T. BEAR, Ph.D.
The plaintiffs called Donald Y.T. Bear, Ph.D., Professor of Economics at the University of California at San Diego. Dr. Bear is an economist with qualifications in statistics.
Dr. Bear analyzed the CDC raw data underlying the Schoenberger study and documented a dropoff in the reporting of GBS cases after December 18, 1976. Dr. Bear assumed that there was complete reporting of vaccinated and unvaccinated GBS cases prior to December 18, 1976. Based on a purely mathematical analysis of the data, he concluded that, after December 18,1976, 26% more unvaccinated GBS cases were reported to CDC than vaccinated GBS cases. Dr. Bear adjusted the CDC raw GBS data based on these assumptions. In making these adjustments, however, he did not consider whether the publicity surrounding the development of GBS in recent vaccinees in the fall of 1976 or medicolegal implications had any effect on the efficiency of reporting GBS cases to CDC.
This court specifically finds that, although Dr. Bear’s mathematical calculations are beyond reproach, the assumptions underlying his calculations are riddled with error, and thus his conclusions are not credible.
Dr. Bear is an economist. He is not a medical doctor, biostatistician, or epidemiologist; nor has he had any training in medicine, biostatistics, or epidemiology. Dr. Bear has no training or experience in evaluating biologic credibility of curves such as the curves of his study, and cannot testify on biologic causation. He has never been involved in an epidemiologic surveillance for disease and has no experience in the design of epidemiologic studies. There is no issue as to Dr. Bear’s ability to perform mathematical calculations. However, he lacks the- specific education, training, and experience necessary to properly evaluate the length of risk of GBS from the swine influenza vaccination. Bear does not know whether his study is consistent with standard epidemiologic principles because he does not know what they are. (117:18— 118:5)
Bear admitted that he is not qualified to evaluate or assess motivational factors in reporting (105:17-106:14), and that he is not competent to assess the behavior of physicians in a medical milieu or assess the manner in which physicians went through the reporting procedure. (113:17-114:12) He has no training in excluding or evaluating sources of bias in an epidemiologic surveillance; he can merely attempt to quantify statistically, as he has attempted to do here. (112:2-15)
Dr. Bear’s two ultimate assumptions are the result of an attempt to reconstruct incomplete data on the incidence of GBS in both vaccinees and non-vaccinees after December 18, 1976. First, he assumes that there was complete reporting of GBS cases in both vaccinees and non-vaccinees whose onset of symptoms was before December 18, 1976. (170:4r-14) Second, he assumes that after that date 26% more of the actual GBS cases in non-vaccinees were reported than the actual GBS cases in vaccinees. (155:14-22)
The source of Dr. Bear’s first assumption — that is, the completeness of reporting of vaccinated
and
unvaccinated cases prior to December 18, 1976 — is the unpublished version of Dr. Langmuir’s paper. (36:5-17) The passage he relies on is as follows:
More interesting is the concurrence of observed and expected cases during the declining weeks
through January.
There is no sign of under-reporting. Apparently cases of GBS following vaccine continued to be reported even though reporting of cases among the unvaccinated was believed to have fallen off.
Plaintiffs’ Exhibit 1, p. 15 (emphasis added). Dr. Langmuir’s paper does not state that there was complete reporting of the vaccinated and unvaccinated cases until December 18, 1976.
Bear’s second assumption forms the basis for his “reporting factor,” the largest manipulation or adjustment of the CDC raw data made by him: Bear assumed that after December 18, 1976, 26% more unvaccinated GBS cases were reported to CDC than vaccinated cases.
It is undisputed that after December 18, 1976, there was an underreporting of GBS cases in both vaccinees and non-vaccinees. However, Dr. Bear has reconstructed data for this crucial period under the assumption that the cases in vaccinees were more underreported than the cases in non-vaccinees.
The reporting factor is Bear’s largest adjustment to the data. By his own admission, it is critical to his analysis to properly estimate his reporting factor. (157:19-158:6) He also admitted that, if there were in reality more efficient reporting of vaccinated GBS eases (the opposite of his assumption), the results of his analysis would be skewed to a higher relative risk for vaccinated GBS cases. (165:5-22)
To account for his belief that unvaccinated cases were reported more efficiently than vaccinated ones, Bear multiplied the reported unvaccinated cases by a lower number (reporting factor) than the vaccinated cases. This calculation increased the attack rate for non-vaccinees by a lesser amount than the attack rate for vaccinees. This, in turn, artificially increased the ratio of the attack rate among vaccinees to the attack rate among non-vaccinees, skewing the results to a higher relative risk for vaccinees.
Bear’s second assumption — that after December 18, 1976, 26% more unvaccinated GBS cases were reported to CDC than vaccinated ones — is epidemiologically unsound and contrary to common sense. Because the hypothesis of the association between the swine influenza vaccine and GBS was known, as was the legal liability of the federal government, and because of the tremendous media publicity surrounding the appearance of clusters of GBS among recent vaccinees, this court specifically finds that there was a greater efficiency of reporting of vaccinated GBS cases. Since more vaccinated GBS cases were reported, Bear’s adjustment for his assumptions regarding underreporting artificially lowered the unvaccinated GBS attack rate and raised the vaccinated GBS attack rate, resulting in an artificially longer and progressively higher period of relative risk.
B. DEFENDANT’S EXPERTS IN EPIDEMIOLOGY
The defendant called Thomas M. Mack, M.D., Professor of Epidemiology in the Department of Pathology and Family and Preventive Medicine at the University of Southern California. The defendant also called W. Alan Hauser, M.D., a neuroepidemiologist and Associate Professor of Neurology at the Gertrude H. Sergievsky College of Physicians and Surgeons at Columbia University. Based on the testimony of these experts, the court rejects the assumptions of Dr. Bear and finds his conclusions unworthy of belief.
Everyone agrees that there was not 100% reporting of GBS cases after December 18, 1976. However, the court agrees with the government’s epidemiological experts and finds that, throughout the period of surveillance, vaccinated cases were less underreported than unvaccinated ones.
The Center for Disease Control is an advisory body having only the ability to request that the states report GBS cases to it. The states reported GBS cases to CDC in their own different styles. The call from CDC to report cases came at a time when the limited resources of state and local
health departments were strained by the swine flu inoculation program itself. CDC received complaints from state health departments that were burdened with time-consuming reporting requirements and were competing for limited resources for other state health department functions. Similarly, the call for surveillance came at a time when CDC was still administering the swine flu program.
Not only did the states look for GBS cases with varying degrees of efficiency, but the physicians reported the cases with varying degrees of efficiency. Some GBS cases were reported through semi-public or public facilities such as large university hospitals, which automatically report cases required to be reported. Other GBS cases were reported by physicians. Some physicians are opposed to reporting cases in general on the basis of physician-patient confidentiality and will not even report cases of venereal disease. The largest number of physicians fall somewhere in between the automatic reporting at large medical centers and those who will not report disease— that is, they report under some circumstances and not under others. The issue then becomes what motivates physicians to report disease.
When the first reports appeared of clusters of GBS in two states, there was a request by CDC to the states to report cases of GBS with onset from October of 1976 on, and presumably a request from the states to the physicians and hospitals in each state to do the same thing. At the same time, there was a tremendous amount of media publicity about the possible association between GBS and the swine influenza vaccination. The GBS-swine flu vaccine question was also discussed with much interest by practicing physicians.
Under the circumstances existing at the time, there was an association in the physicians’ minds between GBS and the swine flu vaccine. The hypothesis CDC sought to test by the epidemiologic surveillance was known by the physicians reporting GBS cases to the state health departments.
Some physicians reported GBS in patients who had been vaccinated in the belief that the patients would derive a long-term benefit from documenting a case that was or may have been due to the government’s error. There was general knowledge that the government had assumed the vaccine manufacturers’ legal liability. Reporting illness in vaccinated persons might help the vaccinated patient’s potential welfare. It is a common experience among epidemiologists that, if something happens to one who has had an immunization, it is much more likely to be reported because the patient had the immunization.
Diagnostic considerations also affected the comparability of the attack rates of GBS in the vaccinated with the unvaccinated. Each state applied its own diagnostic criteria for the GBS eases. Most required only: (a) diagnosis of GBS by a physician and (b) objective evidence of muscle involvement. Some states required diagnosis by a neurologist. No diagnostic criteria were applied to the 1098 cases in the Schoenberger study to ensure that they met acceptable clinical standards for diagnosis of GBS. Applying uniform diagnostic criteria to cases before including them in an epidemiologic study is standard epidemiologic procedure.
Under the diagnostic criteria employed by most states, a number of other neurologic disorders could have been included as GBS cases, including acute myelopathies, botulism, myasthenia gravis, porphyria, acute peripheral neuropathies, neuropathies due to toxins, strokes, and multiple sclerosis.
Mild cases of GBS were much more susceptible to over-diagnosis of GBS and misdiagnosis of GBS. Because there was a higher frequency of mild GBS cases reported after 7 weeks from the start of the swine flu program, the inclusion of non-GBS cases in late weeks had profound effects on the relative risk calculations for the late weeks, skewing them in favor of higher relative risks.
In summary, because of the publicity and various medico-legal implications, the vacci
nated GBS cases were reported with greater relative frequency than the unvaccinated cases. Dr. Bear’s analysis is devoid of credibility because it assumes the opposite of this fact. In addition, the various epidemiologic studies done on the CDC raw data have tended to result in higher and longer periods of relative risk associated with vaccination than is actually the case.
VIII. FINAL CONCLUSIONS
The medical theories advanced by the plaintiffs to show that the swine influenza vaccination caused or contributed to Mr. Padgett’s GBS are speculative and not supported by the medical literature. Dr. Lewis’ sensitization theory lacks a medical and scientific basis and is not generally accepted in the present state of medical and scientific knowledge. Dr. Lewis’ testimony that the plaintiff’s URI could not have been the sole precipitating event of the plaintiff’s GBS because it occurred too close to the onset of the GBS and did not clear before the onset of GBS is contrary to the medical literature.
GBS is thought to be precipitated by various infectious illnesses — most often upper respiratory infections and gastrointestinal infections that occur within four weeks of the onset of GBS. The plaintiff’s GBS was precipitated by his upper respiratory infection, whose onset was ten days before the onset of GBS.
The assumptions underlying Dr. Bear’s analysis are erroneous. There was an underreporting of GBS cases reported to CDC. However, the underreporting of unvaccinated GBS cases was greater in relative frequency than the underreporting of vaccinated cases, the opposite of the assumption made by Dr. Bear.
Dr. Bear’s use of reporting factors based on his assumptions regarding underreporting resulted in longer periods of relative risk being associated with vaccination. Dr. Bear showed an increased risk of GBS in late onset weeks by adjusting his data according to erroneous assumptions regarding the effect of underreporting. His conclusion that there is an increased risk up to eighteen weeks after inoculation conflicts with the medical literature on GBS and lacks biologic credibility.
The analysis and logic employed by Dr. Mack is based on sound epidemiologic principles and experience. The increased incidence of GBS as a result of the swine influenza vaccination ends at most ten weeks after inoculation. Since the plaintiff’s GBS occurred sixteen weeks after his inoculation, the vaccination did not cause or contribute to plaintiff’s GBS.
The plaintiffs have failed to sustain their burden of proving that the swine influenza vaccination administered to John Padgett caused or contributed to his GBS. The Clerk is directed to enter judgment in favor of the defendant, also awarding it the costs of defending this action.