These are appeals arising out of litigation involving the National Swine Flu Immunization Program of 1976, 42 U.S.C. Sec. 247b(j)-(l ) (1976) ("Swine Flu Act"). Plaintiff, Joseph Lima, received a swine flu vaccination on November 12, 1976, during the implementation of the swine flu program sponsored by the United States. Approximately sixteen weeks after the shot, plaintiff was diagnosed as having Guillain-Barre Syndrome (GBS).1 Plaintiff brought this action against the United States for damages allegedly caused by the vaccination. The district court held that plaintiff had failed to establish by a preponderance of the evidence that the swine flu vaccine was a contributing factor or proximate cause of his injuries and dismissed the action with prejudice. In re Swine Flu Immunization Products Liability Litigation, 508 F. Supp. 897 (D.Colo.1981).

Vaccinations were administered nationwide under the program from October 1, 1976 to December 16, 1976. The Swine Flu Act provides that the United States shall be liable with respect to claims submitted after September 30, 1976 for personal injury or death arising from the administration of swine flu vaccine during the swine flu program. 42 U.S.C. Sec. 247b(k)(2)(A). Under the Act, the United States generally is liable in the same manner and to the same extent as it would be in any other action brought against it under the Federal Tort Claims Act. Id.

All actions filed against the United States were consolidated by the Judicial Panel on Multi-District Litigation on February 28, 1978 for coordinated pretrial procedure before the United States District Court for the District of Columbia. The final pretrial order provided, inter alia, that if the United States had stipulated that a plaintiff had GBS, it would remain to be determined whether such injuries were, in fact, caused by administration of the vaccine but that the plaintiff would not be required to establish any theory of liability. Upon completion of the pretrial proceedings, this case was transferred to the United States District Court for the District of Colorado for local discovery and trial.

The trial court's findings of fact with regard to the events leading up to the onset of plaintiff's GBS were as follows: Prior to the inoculation on November 12, 1976 and for three weeks thereafter plaintiff was in excellent health. Beginning approximately on December 1, 1976 plaintiff experienced a general fatigability and loss of energy. On March 1, 1977, plaintiff developed a flu-like illness characterized by diarrhea, fever of 103 degrees, abdominal cramps, runny nose, and chills. The day before being hospitalized, plaintiff experienced blurry vision, numbness of his mouth, and weakness in his legs. On the morning of March 7, 1977, plaintiff's symptoms worsened--his vision was blurred and he developed a lisp in his speech. That afternoon plaintiff visited neurologist Saralee McGroarty, M.D. who immediately hospitalized plaintiff.

Upon admission to the hospital, plaintiff was diagnosed as suffering from GBS. Plaintiff remained hospitalized for almost three months. During that time he was severely paralyzed and underwent surgery to assist his breathing. His respiratory muscles, cranial nerves, and limbs had been paralyzed. At the time of trial, plaintiff had substantially recovered and demonstrated only mild residual effects of GBS.

The government stipulated that plaintiff had GBS. Thus, the only issue to be tried was whether the swine flu vaccination caused plaintiff's GBS. Extensive expert testimony was presented in a non-jury trial by both plaintiff and defendant on the causation issue.

Plaintiff initially called Peter S. Quintero, M.D., a board certified neurologist who was qualified as an expert in neurology. Dr. Quintero testified that in his opinion, plaintiff's swine flue inoculation proximately caused his GBS. In Dr. Quintero's opinion, plaintiff had a "smoldering GBS" which began several weeks after the swine flu vaccination. It was pivotal to Dr. Quintero that plaintiff's fatigability and loss of energy occurred approximately two to three weeks after plaintiff's vaccination and that the exacerbation of plaintiff's illness in March was atypical for GBS. Dr. Quintero concluded that plaintiff was in a hyper-sensitive state prior to the time of the acute stages of his illness. Dr. Quintero admitted on cross-examination that he had never before seen a case of smoldering GBS.

Plaintiff then called Martin Lewis, M.D., Professor and Chairman of the Department of Immunopathology, Loyola University School of Medicine, as an expert in pathology and immunopathology. Dr. Lewis stated that the swine flu vaccination predisposed plaintiff to the subsequent attack of GBS and that the vaccination was a cause of the disorder. Like Dr. Quintero, Dr. Lewis testified that plaintiff's gastroenteritis could not alone have caused GBS because the interval between the gastroenteritis and the GBS was too short. Therefore, it became necessary to look back to the antecedent event of the vaccination in order to interpret the short interval within which plaintiff developed GBS after experiencing gastroenteritis.

Dr. Lewis also testified that he was involved with blood serum testing performed on plaintiff's blood sample. The purpose of these tests was to detect an antibody in the serum which reacted against both the vaccine and the peripheral nerve. The predominate theory regarding the dynamics of GBS is that for some unknown reason the antibodies that occur normally within the human immune system act in a confused way by attacking the covering of the nerves. This results in a breakdown of the myelin sheath (the covering of the nerve) and inhibits the ability of signals to move along the nerves.

The tests showed that the antibody that reacts to the swine flu vaccine also attacks peripheral nerve in some cases. Plaintiff, a laboratory microbiologist, had stored his blood serum over the years thus presenting Dr. Lewis and his colleagues with the unique opportunity to engage in this analysis with reference to GBS. Dr. Lewis testified that plaintiff presented nine serum samples for testing. The nine samples were taken prior to, during, and after plaintiff's episode of GBS. In the samples taken at the height of the disease, there were immune complexes in which there was a cross-reactive antibody against both nerve and antigen (the antigen being the catalyst in the vaccine that incites the antibody). The significance of this positive cross-reactivity, according to Dr. Lewis, is that it indicates a relationship between the vaccine and the GBS. Dr. Lewis admitted that he reached this conclusion based only on tests performed on plaintiff without consideration of any other cases.

Plaintiff also presented the testimony of Charles Poser, M.D., Professor and Chairman of the Department of Neurology, University of Vermont School of Medicine. Dr. Poser examined plaintiff and reviewed plaintiff's medical records covering the period of hospitalization for GBS. Dr. Poser testified that, in his opinion, plaintiff's swine flu shot was the proximate cause of his GBS. The basis for his opinion was that the swine flu shot set up a hypersensitivity within plaintiff's immune mechanism that was subsequently triggered by the viral infection experienced at the beginning of March as manifested by the fever and diarrhea. Dr. Poser also believed that the swine flu vaccination directly produced plaintiff's GBS. Dr. Poser admitted on cross-examination that there were no studies supporting the direct causation theory beyond a ten-week period.

Defendant then presented its case. The first witness called by defendant was Saralee McGroarty, a clinical neurologist qualified as an expert in neurology. It was Dr. McGroarty who examined plaintiff on March 7, 1977 and recommended that he be hospitalized immediately. Dr. McGroarty was asked whether, in her opinion, plaintiff had a smoldering GBS from November to March of 1977. Dr. McGroarty replied that while she had seen many acute cases of GBS, smoldering GBS had no meaning for her.

After Dr. McGroarty, defendant called Steven Ringel, M.D., Associate Professor of Neurology at the University of Colorado Health Center, as an expert in neurology. It was Dr. Ringel's opinion that plaintiff's swine flu vaccination was unrelated to plaintiff's illness. Dr. Ringel stated that, in his opinion, the interval between the shot and plaintiff's GBS was too long for the GBS to be attributed to the shot. Dr. Ringel's testimony revealed that the ten-week limit for indicating a relationship between the vaccine and GBS is an extreme outer limit beyond which there is no indication of any relationship or causal link. Dr. Ringel testified that, typically, people who have GBS will experience a viral syndrome approximately a week or ten days before the onset of the acute paralysis episode as occurred in plaintiff's case. Dr. Ringel did not accept a sensitization theory as postulated by Dr. Poser; in his opinion, that theory is not generally accepted in the neurological community. Dr. Ringel then testified that smoldering GBS is without evidentiary support in the neurological literature. He further testified that, in his opinion, the cause of plaintiff's GBS was the acute gastroenteritis that plaintiff experienced approximately seven days prior to the onset of the GBS.

Defendant also called Peter Kohler, M.D., Professor and Chairman of the Department of Immunology at the University of Colorado Health Center, an expert in immunology. Dr. Kohler testified that he had read the deposition given by Dr. Martin Lewis and that he had seen the results of the tests performed on plaintiff's serum by Dr. Lewis and his associates. Dr. Kohler testified that, in his opinion, the tests carried out by Dr. Lewis did not prove that there is a causal link between the swine flu vaccination and plaintiff's GBS. Dr. Kohler testified that the tests merely show that plaintiff's sera showed antibody against the swine flu and antibody against peripheral nerve. Both of these results would have been expected in anyone who had received the swine flu immunization and who had developed GBS. However, Dr. Kohler indicated that there is no basis to say there was a relationship between the two.

The transcript of the deposition of defendant's witness Robert P. Lisak, M.D., Professor of Neurology at the University of Pennsylvania College of Medicine, was admitted into evidence. With regard to causation, Dr. Lisak testified that plaintiff's swine flu vaccination had nothing to do with his GBS. His opinion was based on his prior experience and the generally accepted view in the medical community that peripheral nervous system diseases such as GBS, arising after immunization of any sort, have an incubation period of no more than three weeks. Dr. Lisak testified that plaintiff's symptoms indicated a classic case of GBS caused by a viral infection.

In addition, Dr. Lisak testified that he was familiar with the tests Dr. Lewis performed on plaintiff's blood. Dr. Lisak did not believe that the test results had any significance in determining causation. Because the only findings by Dr. Lewis were with regard to plaintiff, Dr. Lisak was unable to conclude that the tests showed causation. Dr. Lisak observed that the lack of scientific controls and the numerous uncertainties regarding the data undermined its scientific value on the question of causation.

The trial court engaged in an extensive analysis of the testimony and exhibits presented by both parties. The trial court acknowledged the conflict in the medical testimony concerning the crucial question of a causation, and concluded that plaintiff had failed to establish by a preponderance of the evidence that the swine flu vaccine was a contributing factor or a proximate cause of his Guillain-Barre Syndrome.

On appeal, plaintiff raises the following issues:

(1) Whether the trial court erred in first admitting the expert opinion testimony of Carl Johnson, M.D. an epidemiologist, and later striking it after the close of the evidence;

(2) Whether the trial court erred in placing the burden of proof on the question of causation on plaintiff; and

(3) Whether the trial court erred in certain of its findings of fact (here, plaintiff cites nineteen findings of fact and claims that, taken separately or as a whole, they constitute prejudicial error).

Because the cause of GBS is unknown, both parties placed considerable emphasis on epidemiologic evidence. The principal epidemiologic study is reported in Schonberger, et al., Guillain-Barre Syndrome Following Vaccination in the National Influenza Program, United States, 1976-1977, 110 Am.J. Epidemiology 105 (1979) ("CDC study"). The CDC study consisted of a nationwide surveillance of GBS cases conducted by the Center for Disease Control in Atlanta, Georgia during and after the national swine flu program.

The CDC study presents an analysis of the data received by the CDC on cases of GBS with onset between October 1, 1976 and January 31, 1977. The study analyzed data extending six weeks beyond the end of the swine flu program on December 16, 1976. The number of cases of GBS rose sharply between mid-October and mid-December, peaking the week ending December 18, 1976. Thereafter, the incidence of GBS decreased rapidly and eventually leveled out at a point approximately equal to the normal occurrence for GBS in the unvaccinated population.

The data further revealed that the greatest risk of GBS occurred in the second and third weeks after vaccination. The risk remained significantly above the norm up to the tenth week after vaccination. After the tenth week, there was no significantly greater risk of contracting GBS. The authors of the CDC study concluded that there is no relation between swine flu vaccine and the onset of GBS more than ten weeks after vaccination.

One of plaintiff's primary arguments in support of causation was that statistical data regarding the number of cases of GBS in Colorado showed an increase above the norm in March of 1977 (the month in which plaintiff was diagnosed as having GBS). To compile these statistics, plaintiff undertook his own survey of Colorado hospitals. Plaintiff sought information regarding the number of GBS cases treated from approximately 1976 to 1978, the onset date of each case, the hospital admission date and whether the patient had been vaccinated. Plaintiff also obtained statistical data on GBS cases from the Colorado Department of Health. The result of these compilations was plaintiff's Exhibit 57, which consisted of the replies from the hospitals and plaintiff's summary of the data.

Carl Johnson, M.D., a specialist in epidemiology, testified on behalf of plaintiff's statistically-based analysis. Defendant objected on the ground that Dr. Johnson's testimony would be based in part on plaintiff's Exhibit 57, which had just been provided to the defendant on the morning of the trial. The court decided, in response to this objection, that it would hear the testimony, reserving any ruling on whether the testimony would be admitted. The court indicated that the testimony of Dr. Johnson might be stricken at a later time.

Exhibit 57 consists primarily of letters from medical records technicians of hospitals in Colorado regarding the number of cases of GBS that each hospital had treated from 1976 to 1978. This information was then summarized by plaintiff's attorney to determine whether there had been an increase in the number of GBS cases in March of 1977.

Dr. Johnson testified that, in his opinion, the underlying data found in Exhibit 57 were of a type which would be reasonably relied upon by an expert in the field. The data contained in Exhibit 57, according to Dr. Johnson, indicated that there had been an increase in GBS cases in Colorado in March of 1977. Dr. Johnson further testified that the information received for preparation of Exhibit 57 was based on the International Classification of Diseases Adapted, which designates a specific code for various types of illnesses. The code which would include GBS is commonly referred to as ICDA Code 354. In deciding what to report in response to plaintiff's request, medical technicians apparently relied on the ICDA Code 354.2

Dr. Johnson admitted on cross-examination that he did not evaluate the length of time between immunization and the onset of GBS. Indeed, an examination of Exhibit 57 reveals that the date of the vaccination is unknown in certain instances. Dr. Johnson further acknowledged that the exhibit contained cases that (a) were not GBS; (b) were counted twice; (c) were from persons inoculated in states other than Colorado; (d) had questionable diagnoses; and (e) were recorded as being during a time when the national moratorium on the immunization was in effect.

At the close of Dr. Johnson's testimony, defendant objected to the entire testimony as being based on Exhibit 57, which was inherently unreliable and not of the type reasonably relied on by experts in the field. The court overruled defendant's objection and permitted the testimony to stand.

Later in the trial, during presentation of defendant's case, Steven Ringel, M.D., testified that plaintiff's Exhibit 57 contained cases of illnesses other than GBS. Dr. Ringel stated that Exhibit 57 was not the type of information that he, as an expert in neurology, would rely on in formulating an opinion. According to Dr. Ringel, the information contained in plaintiff's Exhibit 57 was provided by a clerk rather than a neurologist. These clerical personnel did not consider the clinical picture, nor were they qualified to do so, in deciding whether all of the cases were in fact GBS. Dr. Ringel noted that the responses from the various clerks indicated that some cases were possible GBS, some were polyneuropathies, and that there was uncertainty with regard to these cases as to whether they were GBS at all.

After the trial, defendant renewed its objection to Dr. Johnson's testimony. The trial court granted defendant's motion to strike on the basis that Dr. Johnson's opinion did not reach the threshold standards for admissibility as set forth under Fed.R.Evid. 703. The trial court found that the data (Exhibit 57) upon which Dr. Johnson relied were inadmissible hearsay and were "not of the type reasonably relied on by experts in epidemiology and neurology." In re Swine Flu Immunization Products Liability Litigation, 508 F. Supp. at 904.

Plaintiff contends on appeal that it was an abuse of discretion for the trial court to strike Dr. Johnson's testimony after previously admitting it. It is plaintiff's position that additional evidence was available to buttress the data contained in Exhibit 57 and would have been presented had the trial court not initially admitted Dr. Johnson's testimony. Plaintiff contends that it is fundamentally unfair for crucial evidence to be deemed admissible during the trial and then stricken when the time for presenting additional evidence has passed.

The issue, therefore, is not whether the trial court's ruling was proper from an evidentiary standpoint, for even plaintiff acknowledges that "[i]t may have been error to admit Dr. Johnson's testimony in the first place." Initial Brief for Appellant at 30. The question on appeal is whether prejudicial error lies in the striking of the "crucial evidence" after the court had initially ruled it admissible.

We are persuaded, however, that the testimony of Dr. Johnson is not crucial evidence in this case. The trial court concluded its discussion on the evidentiary ruling with the following statement. "[A]ssuming [Dr. Johnson's testimony] is admissible, the testimony has limited persuasive effect. It clearly does not tip the scales in favor of causation." 508 F. Supp. at 904. This indicates that the trial court did consider and evaluate the testimony. The trial court's factual findings are presumptively correct and the appellate court will not disturb them unless they are clearly erroneous. Fed.R.Civ.P. 52(a); Gutierrez v. Denver Post, Inc., 691 F.2d 945, 946 (10th Cir.1982). In order to conclude that the trial court's findings are clearly erroneous, we must be "left with the definite and firm conviction that a mistake has been committed." Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 123, 89 S. Ct. 1562, 1576, 23 L. Ed. 2d 129 (1969) (quoting United States v. United States Gypsum Co., 333 U.S. 364, 395, 68 S. Ct. 525, 542, 92 L. Ed. 746 (1948)). Moreover, the appellate court is to give due regard to the "opportunity of the trial court to judge the credibility of the witnesses." Fed.R.Civ.P. 52(a). There is substantial evidence in the record to support the district court's finding of "limited persuasive effect" with regard to Dr. Johnson's testimony. In light of the trial court's careful consideration of the testimony, we cannot say that its finding is clearly erroneous.

Plaintiff's argument for a shift in the burden of proof on the question of causation is premised upon the fact that medical science does not know what causes GBS. Given that unknown, plaintiff contends that the government is better equipped to provide statistical data than is the individual plaintiff. Therefore, in order to provide a meaningful remedy, plaintiff should be entitled to a presumption of causation so defendant will come forward with the available evidence. Plaintiff draws analogies to various negligence cases involving shifts of the burden of proof from plaintiffs to defendants.

The remedial provisions of the Swine Flu Act state in part as follows:

The United States shall be liable with respect to claims submitted after September 30, 1976 for personal injury or death arising out of the administration of swine flu vaccine under the swine flu program and based upon the act or omission of a program participant in the same manner and to the same extent as the United States would be liable in any other action brought against it under such section 1346(b) and chapter 171 [Federal Tort Claims Act, "FTCA"] except that--

(i) liability of the United States arising out of the act or omission of a program participant may be based on any theory of liability that would govern an action against such program participant under the law of the place where the act or omission occurred, including negligence, strict liability in tort, and breach warranty ....

42 U.S.C. Sec. 247b(k)(2)(A). Congress has expressly directed the procedure for the handling of claims for personal injury arising out of the administration of the swine flu vaccine. An examination of the legislative history of the Swine Flu Act clearly reveals that Congress did not intend any change in the normal application of the Federal Tort Claims Act other than in specifically expressed instances not pertinent here. See 122 Cong.Rec. 26,62 5-40, 26,793-813 (1976).

In the present action, since the vaccination occurred in Colorado, Colorado law governs this issue. 28 U.S.C. Sec. 1346(b); Richards v. United States, 369 U.S. 1, 82 S. Ct. 585, 7 L. Ed. 2d 492 (1962). The trial court, citing Exchange National Bank v. Sparkman, 191 Colo. 534, 554 P.2d 1090, 1092 (Colo.1976), concluded that plaintiff has the burden of proving by a preponderance of the evidence that the swine flu vaccine was the proximate cause of his GBS. The court held that in order for the swine fluvaccine to be a proximate cause of plaintiff's GBS, it must be shown the illness would not have occurred but for the inoculation or that the inoculation was a contributing factor bringing about the disease. Reaves v. Horton, 33 Colo. App. 186, 518 P.2d 1380 (1973), rev'd on other grounds, 186 Colo. 149, 526 P.2d 304 (1974). We agree.

Many of the authorities cited by plaintiff are not from Colorado. The direction provided by the Swine Flu Act as well as the Federal Tort Claims Act requires us to consider the law of Colorado. We address the arguments raised by plaintiff insofar as they are based on Colorado law. Among the analogies suggested, plaintiff likens the instant case to the doctrine of res ipsa loquitur. In order to apply this doctrine in Colorado, it is necessary to meet the essential elements as set forth in Montgomery Elevator Co. v. Gordon, Colo., 619 P.2d 66 (Colo.1980). In the instant case, the first essential element in Montgomery is dispositive: that the event [plaintiff's GBS] is the kind that ordinarily does not occur in the absence of negligence. Id. 619 P.2d at 68. It is well accepted in the medical community that GBS can occur in the absence of negligence, indeed it can occur without any precipitating event. Plaintiff's Exhibit 26, Arnason, Inflammatory Polyradiculoneuropathies, MDL 330 at 5 (50% of GBS cases have no obvious antecedent event).

Plaintiff also relies on Prutch v. Ford Motor Co., Colo., 618 P.2d 657 (Colo.1980). Prutch dealt with products liability and the allocation of the burden of proof on the question of the point at which a defect appeared in a product in the manufacturing-delivery process. Id. at 660. The Colorado Supreme Court concluded that requiring each defendant in the chain of distribution to show that a product was not defective when it left its control poses no unreasonable burden on defendants. Such a procedure simply redistributes the burden to those who have superior knowledge of the truth and better access to evidence. Id. The Colorado Supreme Court's allocation of the burden of proof in Prutch clearly depends on the relative ease of the defendant's accessibility to evidence. In the instant case, however, knowledge of causation is equally inaccessible to both plaintiffs and defendants, and both have access to the same studies and information. We conclude that under the law of Colorado, this case does not present an instance where the burden of proof can be shifted on the question of causation to the defendant.3 We hold further that the trial court applied the proper standard on the burden of proof.

In his final challenge to the district court's decision, plaintiff contends that nineteen miscellaneous factual findings taken from the opinion of the district court are erroneous and prejudicial. Findings of fact will not be overturned on appeal unless they are clearly erroneous. Fed.R.Civ.P. 52(a). A finding fact is "clearly erroneous" if this court, upon review, is left with a definite and firm conviction that a mistake has been made. Gutierrez v. Denver Post, Inc., 691 F.2d 945, 946 (10th Cir.1982) (and cases cited therein). It is not necessary to review each finding here. Suffice it to say, we have considered the record of the district court's findings and we are not left with the requisite firm conviction that a mistake has been made.

Accordingly, the judgment of the district court dismissing plaintiff's action with prejudice is hereby affirmed.

Plaintiff has also appealed the order of the district court denying plaintiff's motion for relief from judgment based on allegedly newly-discovered evidence (Appeal No. 81-1640). Finding no abuse of discretion in the district court's order, we hereby affirm.