Griffin v. United States, 351 F. Supp. 10 (E.D. Pa. 1972)
November 7, 1972
United States District Court, E. D. Pennsylvania.
*11 *12 *13 Avram G. Adler, and Stanley P. Kops, Freedman, Borowsky & Lorry, Philadelphia, Pa., for plaintiff.
John Charles Kruse, and Robert D. Batson, Dept. of Justice, Washington, D. C., for defendant.OPINION AND ORDER
NEWCOMER, District Judge.
On October 27, 1963, plaintiff, Mary Jane Griffin, participated in a mass campaign to eradicate the dreaded disease of poliomyelitis sponsored by the Montgomery County Medical Society, by swallowing a dose of Sabin live virus oral polio vaccine. Just over a month later, Mary Jane Griffin lay near death in a coma at Lankenau Hospital in Philadelphia. She emerged from that coma a permanent quadriplegic. In this suit, it is urged that the United States is liable to Mrs. Griffin and her husband, Richard Griffin, for her tragic condition under the Federal Tort Claims Act, 28 U.S.C. §§ 1346(b), 2671 et seq.
This action was commenced on October 26, 1965. It was tried to the Court without jury from July 10 to July 27, 1972. In accordance with rule 52(a) of the Federal Rules of Civil Procedure, this Court makes the following findings of fact and conclusions of law:FINDINGS OF FACTLIABILITY
1. Plaintiffs, Mary Jane Griffin and her husband, Richard J. Griffin, are both citizens of the Commonwealth of Pennsylvania.
2. At all times pertinent to this action, both plaintiffs have lived together at 317 Conshohocken State Road, Gladwyne, Montgomery County, Pennsylvania.
3. At the time of trial, plaintiff Mary Jane Griffin was 50 years old.
4. At the time of trial, plaintiff Richard J. Griffin was 54 years old.
*14 5. The Griffins have two children, Mary and Jane, who at the time of trial were 15 years old and 19 years old respectively.
6. Gladwyne, Pennsylvania is an upper socio-economic suburb of the City of Philadelphia, Pennsylvania.
7. In the year 1963, Mary Jane Griffin was primarily a housewife taking care of her husband and two children.
8. In the summer of 1963, Mary Jane Griffin and her two daughters visited with her parents in Longport, New Jersey until after Labor Day when they returned to their home in Gladwyne, Pennsylvania.
9. In addition to her duties as wife and mother in the year 1963, Mary Jane Griffin was employed by the Atlantic City Race Track of Atlantic City, New Jersey, in a part-time capacity to do public relations work for the Race Track involving the staging of fashion shows on Tuesdays during its summer meeting.
10. Mrs. Griffin's duties at the Atlantic City Race Track were concluded by the end of September, 1963.
11. Mrs. Griffin did not leave the Gladwyne area between September 1963 and the onset of her symptoms in November of 1963.
12. Beginning June 1, 1963, cases of poliomyelitis, Type I, began to appear in certain sections of the City of Chester and the City of Philadelphia.
13. The Montgomery County Medical Society, spurred by the appearance of these cases, decided to sponsor a mass campaign of immunization against all three known types of polio, using the Sabin live-virus oral polio vaccine.
14. The campaign was to be conducted, and ultimately was conducted in Montgomery County, Philadelphia County, Delaware County and Bucks County, Pennsylvania, and Camden, New Jersey.
15. The contract to supply the campaign with live-oral polio vaccine went to Charles Pfizer Company, the American distributor of Sabin live-oral polio vaccine manufactured by Pfizer, Ltd. of Sandwich, England, a subsidiary of the American Corporation. (Both entities are hereinafter referred to simply as "Pfizer").
16. The contract called for Pfizer to supply the campaign with 500,000 dosages of each of the three types of Sabin vaccine, Type I, Type II and Type III.
17. Mary Jane Griffin, participating in the mass campaign, took Sabin Type I vaccine at the Gladwyne School on September 22, 1963, along with the other members of her family.
18. Mary Jane Griffin, participating in the mass campaign, took Sabin Type III vaccine at the Gladwyne School on October 27, 1963, along with the other members of her family.
19. The vaccine used in the October 27 feeding at Gladwyne School was taken from an undetermined number of doses of Pfizer Lot 71 Type III vaccine and an undetermined number of doses of Pfizer Lot 56 Type III vaccine provided to Montgomery County by Pfizer for use in the campaign.
20. As a result of taking this vaccine, Mary Jane Griffin developed Type III poliomyelitis, which is responsible for her present paralysis.
21. Mary Jane Griffin ingested a dose of Pfizer Lot 56 oral polio vaccine.
22. Pfizer Lot 56 was approved and released for sale and human consumption by the Division of Biologic Standards of the National Institutes of Health, a part of the Department of Health, Education and Welfare, hereinafter referred to as D.B.S., in Bethesda, Maryland by letter of July 9, 1962.
23. The testing and evaluation which led to the release of Lot 56 was done by D.B.S. in Bethesda, Maryland.
24. The regulation governing the evaluation of monkey neurovirulence testing at all times pertinent to this action was 42 CFR 73.114(b) (1) (iii) promulgated *15 March 25, 1961, which reads as follows:Determination of neurovirulence. At the conclusion of the observation period comparative histopathological examinations shall be made of the lumbar cord, cervical cord, lower medulla, upper medulla and mesencaphalon of each monkey in the groups injected with virus under test and those injeced with the NIH Reference Attenuated poliovirus, except that for animals dying during the test period, these examinations shall be made immediately after death. The animals shall be examined to ascertain whether the distribution and histological nature of the lesions are characteristic of poliovirus infection. A comparative evaluation shall be made of the evidence of neurovirulence of the virus under test and the NIH Reference Attenuated Poliovirus with respect to (a) the number of animals showing lesions characteristic of poliovirus infection, (b) the number of animals showing lesions other than those characteristic of poliovirus infection, (c) the severity of the lesions, (d) the degree of dissemination of the lesions, and (e) the rate of occurrence of paralysis not attributable to the mechanical injury resulting from inoculation trauma. The virus pool under test is satisfactory for poliovirus vaccine manufacture only if at least 80 percent of the animals in each group survive the observation period and if a comparative analysis of the test results demonstrate that the neurovirulence of the test virus pool does not exceed that of the NIH Reference Attenuated Poliovirus." (Emphasis added)
25. The reference strain for monkey neurovirulence testing was a Type I strain called NA-2.
26. The reason for monkey neurovirulence testing of a live oral polio vaccine lot was to attempt to insure that the lot was safe to those who ingested vaccine from the lot.
27. On July 9, 1962, the personnel of D.B.S. charged with evaluating the results of the monkey neurovirulence tests knew that Sabin Type III vaccine virus was more genetically unstable than NA-2 on in vivo passage and might be genetically unstable on in vitro passage.
28. On July 9, 1962, the personnel of D.B.S. charged with evaluating the results of the monkey neurovirulence tests knew that the neurovirulence characteristics of Sabin Type III virus had changed in a way suggestive of possible danger after the removal of Simian virus 40 from the original seed pools, that is, that Sabin Type III then produced more severe lesions in the intrathalamic test.
29. On July 9, 1962, the personnel of D.B.S. charged with evaluating the results of the monkey neurovirulence tests had accepted the proposition that the intrathalamic test was the most important and critical test of monkey neurovirulence.
30. On July 9, 1962, the personnel of D.B.S. charged with evaluating the results of the monkey neurovirulence tests knew that the test results of the manufacturing laboratories submitting protocols of their own monkey neurovirulence testing were unreliable and could not be given weight in the decision to approve and release a given vaccine lot.
31. At the time the protocol for Pfizer Production Lot 56 was submitted to D.B.S. for approval, the cumulative experience of D.B.S. for NA-2 concerning lesion scores as a result of intrathalamic inoculation was based on 305 monkeys. From the standpoint of severity, which is the average lesion score of the site closest to inoculation, only one monkey with an average Grade 3 lesion had been observed; from the standpoint of spread, which is the average lesion score of the sight furthest from the inoculation, only one monkey with an average Grade 3 lesion had been observed. All other lesions were either Grade 1 or Grade 2 lesions. No monkeys showed paralysis by clinical observation.
*16 32. Lesions are scored in the following manner: Grade 1, minimal; Grade 2, mild; Grade 3, moderately severe; and Grade 4, severe.
33. The minimal lesion, a Grade 1 lesion, is where there is destruction of one or two anterior horn cells on one side of the spinal cord in one horn. A mild lesion, a Grade 2 lesion, is where there is destruction of more than two but less than five anterior horn cells on one side or destruction of one or two anterior horn cells on both sides. A moderately severe lesion or a Grade 3 lesion, is where only one or two anterior horn cells remain undamaged on both sides. A Grade 4 lesion, or a severe lesion, is where all anterior horn cells on both sides are totally destroyed.
34. Production Lot 56, when tested by D.B.S. for monkey neurovirulence by inoculation of thirty monkeys, demonstrated that four monkeys showed lesions. As to severity of lesions, two monkeys showed average Grade 4 lesions. One monkey showed average Grade 3 lesions. One monkey showed average Grade 2 lesions. For spread, three monkeys showed average Grade 3 lesions and one monkey showed average Grade 2 lesions. One of the monkeys showing Grade 4 lesions also demonstrated paralysis not related to the trauma of inoculation by clinical observation.
35. At the time that D.B.S. completed their monkey neurovirulence testing, the cumulative experience of lesion scores as a result of D.B.S.'s intrathalamic inoculations of 305 monkeys with reference strain NA-2 demonstrated that there had never been a paralyzed monkey and that there had never been any monkeys with average Grade 4 lesions either as to severity or spread.
36. Out of the 305 monkeys intrathalamically inoculated by D.B.S., 24 monkeys showed lesions regardless of Grade. On an average basis per 30 monkeys this would be the equivalent of 2.36 lesions per 30 monkeys inoculated. Of the 24 monkeys showing lesions, 13 monkeys showed only minimal lesions which by definition were lesions affecting only one or two horn cells.
37. Production Lot 56 showed four monkeys with lesions, all of which exceded minimal grade scores.
38. D.B.S. experience with NA-2 showed that NA-2 viewed in terms of 30 monkey lots produced four lesions of any kind in only about ¼ of the cases (4 of 15 test groups by defense exhibit 98).
39. Lot 56 fell into the upper range of experience with NA-2 as to incidence of lesions both viewed in terms of the average experience with NA-2 and the experience with NA-2 in 30 monkey lots.
40. A comparative analysis of the test results obtained in testing Lot 56 for monkey neurovirulence and the NA-2 experience did not demonstrate that Lot 56 did not exceed the reference in neurovirulence.
41. A comparison of the test results obtained in testing Lot 56 for monkey neurovirulence and the NA-2 experience demonstrated that Lot 56 probably exceeded the reference in neurovirulence.
42. On July 9, 1962, the personnel of D.B.S. charged with evaluating the results of the monkey neurovirulence test knew or should have known that Lot 56 probably exceeded the reference in neurovirulence.
43. On July 9, 1962, the personnel of D.B.S. charged with evaluating the results of the monkey neurovirulence test definitely knew or should have known that a comparison of the test results to the cumulative experience with NA-2 did not demonstrate that Lot 56 did not exceed NA-2 in neurovirulence.
44. All information received by D.B. S. between July 9, 1962 and October 22, 1963 indicated that Lot 56 did in fact exceed the reference strain in neurovirulence.
45. D.B.S. could have recalled Lot 56 at all times between July 9, 1962 and October 22, 1963.
46. D.B.S. did not recall Lot 56 between July 9, 1962 and October 22, 1963.
*17 47. It was negligent for D.B.S. to release Lot 56.
48. It was negligent for D.B.S. not to recall Lot 56.
49. This negligence imposed a higher risk on the group who were intended to be protected by the monkey neurovirulence testingthose who ingested Lot 56 oral polio vaccine.
50. Harm to some members of this group was a foreseeable consequence of this negligence.
51. This potential risk imposed by the negligence of D.B.S. came to fruition in the case of Mary Jane Griffin.
52. Mary Jane Griffin's disease was caused by the negligence of the Division of Biologic Standards.DISCUSSION
The following discussion is limited to findings of fact about which there is substantial controversy.
It should be noted at the outset that the Court is somewhat handicapped in determining the true state of the facts concerning what exactly was swallowed by Mrs. Griffin. A few things are clear. First, the plaintiffs have never had custody of any records other than those now in court which might throw light on this question. Second, Pfizer has in the past had such records in some detail and did not produce them in response to plaintiffs' subpoena (e. g. the lot record cards mentioned in plaintiffs' exhibit 31). As to finding 19, that doses of both Lot 56 and Lot 71 were used in the Montgomery County program is established clearly by plaintiffs' exhibit 31 A-K. These 11 photocopies of letters and inter-office memoranda from late 1965early 1966, are everything Pfizer could produce on the subject, although from 1965 until as recently as just over a year ago, Pfizer was the defendant in a case arising out of the identical subject matter. All of the documents agree that 56 and 71 both went to Montgomery County. These include interoffice memos and letters to Pfizer's own counsel. All these documents except one, however, are silent on the relative quantities of 56 and 71 supplied. That single document, exhibit D-92, is a photocopy of a scribbled memo on the note paper of C. F. Hagan, Asst. General Counsel to Pfizer. The memo itself is undated and unsigned. The paper it is mounted on, apparently for purposes of photocopying, bears the inscription "Rec'd in file, 12/9/65." The memo was prepared with the suit against Pfizer in mind, as evidenced by the annotation at the top of the memo. The name of Will Donaldson is written at the top. Whether he wrote the memo, gave the information found thereon, or was to be the recipient thereof, is unclear. The government contends that this document establishes that 5500 vials of Lot 56 were sent to Montgomery County. The government further contends that, once this issue of the relative quantities of 56 and 71 is raised by the introduction of evidence, the burden of persuasion as to that issue is on the plaintiffs. The Court disagrees with both propositions.
First, it is true that plaintiffs bear the burden of persuasion on each fact necessary for their case. Under certain circumstances, that plaintiff contracted polio from a dose of Lot 56 vaccine could be such a fact. There are no direct proofs of this fact and it must be proved by circumstantial evidence. *18 Plaintiffs presented sufficient evidence of the presence of Lot 56 in Montgomery County, the nature of Lot 56, and the nature of plaintiff's disease to support such a conclusion. The defense then sought to establish a circumstance tending to make that conclusion less likely, though not inevitably untrue, i. e. the purported ratio of Lot 71 vaccine to Lot 56 vaccine in Montgomery County. While the trier of fact must be persuaded by a preponderance of all relevant evidence that plaintiff in fact ingested a dose of Lot 56, the defense must necessarily prove any specific circumstance it alleges to be true tending to reduce if it wishes the trier of fact to regard such circumstance as established. The difference is the difference between a fact ultimately necessary to plaintiffs' case, and a circumstantial fact bearing on but not dispositive of such ultimate fact issue. As to such a circumstantial fact, he who urges it must prove it to gain the benefit of its being accepted as true.
The evidence the government adduced fails to prove much of anything to this Court's mind. The Government argues that viewing Exhibit D-92 together with the invoices produced by the Montgomery County Medical Association by which Pfizer billed the Montgomery County Medical Society for the vaccine it was sent for the 1963 mass immunization campaign (D-95) compels the conclusion that D-92 is accurate and further that no Lot 56 vaccine was in Montgomery County on October 27 when Mrs. Griffin got her dose. In the opinion of this Court a comparison of the two exhibits indicates the exact opposite on both counts.
The Government reasons as follows: First, D-92 says that 5500 vials of Lot 71 and 30 vials of Lot 56 went to Montgomery County. Second, D-92 says that Type III vaccine was shipped to Montgomery County on October 21, 1963 and on November 19, 1963. Third, invoice number 878155 dated October 24, 1963 shows the shipment of 550 boxes of ten 100 dose vials of type III vaccine (5500 100 dose vials). This, the government urges, was the Lot 71. No other invoice produced evidences the shipment of any other Type III vaccine, but one invoice is obviously missing. Both the November 26 remittance stub and the December 26 remittance stub show an invoice number 17362 dated November 19, 1963 in the amount of $48.30 which was not produced. This missing November 19 invoice, the government argues, taken with the November 19 shipping date for some Type III vaccine alleged in exhibit D-92, must represent the 30 vials of Lot 56. Hence, no Lot 56 vaccine was in Montgomery County on October 27 when Mrs. Griffin took her sugar cube.
The problems with the government's argument are numerous. First, the dates on the invoices seem to bear no relation to actual shipping dates. Thus, invoice B73088, which represents virtually the entire shipment of Type I vaccine, is dated November 22, 1963, but we know from all the testimony on the matter that the Type I was actually administered long before this date, on September 22 and September 29, and even D-92 says that type I was shipped in September. Further, D-92 says that Type III was shipped on October 21, but the invoice already alluded to is dated October 24. Given this, one would hardly be *19 compelled to conclude that the missing November 19 invoice represents a November 19 shipment of type III vaccine such as the one alleged in D-92. If anything, the opposite is indicated.
Second, it is peculiar that, of all the Lot quantities given in D-92 and the various invoices for different vaccine types, only the figure for Type III Lot 71 corresponds exactly to anything in the invoices. Thus D-92 shows a total of 5500 vials of Type I shipped to Montgomery County, while the invoices reveal only 4334 plus 4 separate 10 dose vials. It is true that there is a pencil notationon invoice B73088 which says 4313 1186 ____ 5500
but even speculating that this represented some unaccounted gift of 1186 vials by Pfizer, the total on D-92 is still 20 vials and 40 doses off, which doesn't say much for D-92 as evidence of exact quantities. And this assumption about the phantom 1186 vials of Type I would be pure speculation. The 1186 vials appear nowhere on the November or December statements. Further, although if one takes the illegible second digit of Type II, Lot XXX-XXXX-X in D-92 to be a three, the quantities of Type II on D-92 and Type II in the invoices correspond, it should be noted that this means that Type I and Type II were shipped in shipments from many varied lots. Only for Type III do we get the suggestion that one invoice represents all and only all of one lot. But what about the 30 vials of Lot 56? The missing November 19 invoice wouldn't account for them, even if such a late shipment of a significant quantity of Type III vaccine were credible in itself. That bill totaled $48.30, which is the price of less than 10 vials of the vaccine (all vaccines sold at $49.60 per ten vials). Whatever type vaccine was included, there were not more than nine 100 dose vials, plus possibly some smaller quantities. And once again, the totals of D-92 and the invoices do not square, for D-92 shows 5530 vials of Type III, and the invoices reveal not more than 5509 by any speculation.
Further, all figures given above taken from the invoices represent the amount of vaccine received, but we know that 1290 of a total of 5500 vials of type III were returned for credit, leaving only 4219 vials used. (Similarly, 952 out of 5,000 vials of Type II were returned, leaving only 4048 used.) This fact raises an interesting question. Why did whoever wrote down the figures on D-92 concern himself with the quantities of vaccine lots which went into Montgomery County when the relevant question was what lots were used, and fully 1/5 of the amount shipped in was shipped back unused for credit? If D-92 was abstracted first hand from the lot distribution record cards in 1965, it is almost unthinkable that this information was not apparent.
Four conclusions seem to come from a comparison of D-92 with the invoices.
1. Invoice dates don't seem to be shipping dates.
2. Most Pfizer shipments were comprised of various lots.
3. Even with the utmost allowable speculation, the figures for total shipment to Montgomery County on D-92 don't add up to the figures on the invoices in the case of Type I vaccine and Type III vaccine.
*20 4. The missing November 19 invoice did not represent 30 vials of Lot 56 vaccine.
The Court has concluded that D-92 is untrustworthy for the following reasons. Many of the reasons stated would ordinarily render the note inadmissible. It was admitted by stipulation of counsel, as were hundreds of other documents, but this does not mean that the Court must ignore these factors bearing on the note's probative value. Of all the documents admitted by stipulation, only this one stands out sharply as markedly untrustworthy.
1. It is a self-serving declaration prepared by some agent of a defendant in a law suit on the same facts.
2. The author is unknown.
3. The source of the information is unknown.
4. The date is unknown.
5. The hurried nature of the writing raises questions of copying accuracy which cannot be checked.
6. The original lot record cards which would show the information accurately were not produced by Pfizer though they must have been kept until 1967 by 42 CFR 73.37 a & b, and until 1971 pending Pfizer's settlement with plaintiffs. Pfizer knew of the present action in 1965, knew that those records would be called for, and, it must be assumed, would have held them ready and produced them if they reflected well for Pfizer. Though the government cannot be charged with Pfizer's acts, they must accept any inferences of credibility and trustworthiness which naturally flow from those acts. It is frankly incredible that Pfizer possessed only 11 photocopies of rather cursory documents bearing on this case.
7. The figures on the note cannot be reconciled with the invoices prepared by Pfizer and brought in by the Montgomery County Medical Society.
Having concluded that the only evidence of the relative quantities of Lot 71 and Lot 56 adduced at trial is untrustworthy, this Court can only assume that the quantities actually present during the October 27 feeding were relatively equal. However, even if this Court were convinced that there was a greater quantity of Lot 71 than Lot 56, we would still conclude that Mrs. Griffin's dose was from Lot 56 for reasons apparent from the discussion that follows:
First, there is no doubt in the Court's mind that Mrs. Griffin had Type III poliomyelitis and that her illness was caused by whatever dose of Type III vaccine she ingested. It may be taken as established virtually beyond doubt that some Type III Sabin live-oral vaccine can give some persons paralytic polio. By July of 1964, 15 vaccine related cases deemed compatible with a conclusion *21 of vaccine causation had followed Type I feedings, 2 had followed Type II feedings and 36 had followed Type III feedings. The Surgeon General's own advisory committee on live polio virus vaccine calculated the probability of the ratios occurring randomly if there were in fact no risk or equal risk among the three vaccine types at 1 in 185 million. The Committee stated clearly that in its view "at least some of these cases were caused by the vaccine."
The Government suggested that Mrs. Griffin might coincidentally have contracted some other paralytic disease, either Guillaine Barré syndrome, an ECHO virus, or a Coxsackie virus. The biggest flaw in this proposition is that the entire world literature on these three possible alternative sources of Mrs. Griffin's illness fails to reveal one substantiated case of permanent quadriplegic paralysis attributable to such cause. All three of these possible alternative causes induce transient paralysis only. Further, there was no epidemological data which would indicate that any of these agents were around the Montgomery County area in virulent form in the Fall of 1963. Mrs. Griffin's illness was not Guillaine Barré Syndrome, or caused by an ECHO or Coxsacki virus. It was, from her symptoms, poliomyelitis.
Further, it could not have been vaccine-induced Type I polio, since the time between her Type I feeding on October 22, and the onset of her symptoms was beyond the incubation period for poliomyelitis. Further, her disease was not wild Type I virus, since her Type I feeding protected her against wild Type I virus after feeding, and any pre-existing wild Type I infection would have manifested itself symptomatically before the actual onset of symptoms. This leaves us with three possibilitieswild Type II, wild Type III, or vaccine induced Type III (Mrs. Griffin took no Type II vaccine). There appears to have been no virulent wild Type II or Type III virus around Montgomery County when Mrs. Griffin became ill. Indeed, no Type II cases were reported in the five Pennsylvania county area of Philadelphia, Montgomery, Bucks, Delaware and Chester Counties from August 1 to December 31, 1963 at all, and the only Type III cases reported were also vaccine associated. The absence of wild Type II virus, the probable absence of wild Type III virus, and the perfect correspondence between the ingestion of Type III vaccine from a vaccine universe known to contain at least some vaccine from a lot, lot 56, which could induce paralytic poliomyelitis in some subjects (see discussion below) leads this Court to the confident conclusion that Mary Jane Griffin had Type III paralytic poliomyelitis which she contracted from a dose of oral polio vaccine taken on October 27, 1963.
This Court has concluded that Mrs. Griffin's dose was a dose of Lot 56 basically on three groundsFirst, the high monkey neurovirulence of Lot 56, discussed at length below; Second, the association of Lot 56 with other cases; and third, the lack of association of Lot 71 with other cases. The government contends that evidence of monkey neurovirulence cannot be extrapolated to expectations of virulence in man. The Court disagrees. The relevant data seem to indicate the opposite conclusion, at least on a gross scale. Thus Sabin Type II showed up consistently lower quantitatively and qualitatively on the Monkey Neurovirulence tests than did the NA-2 reference strain, a Type I strain, and Type II vaccine was associated with very few cases. Sabin Type I was higher *22 on the average than Type II and was associated with more cases. The original Sabin Type III with SV-40 was lower than NA-2 and the clinical field trial experience, such as it was, did not indicate the kind of human risk later encountered in Sabin Type III without SV-40. Once SV-40 was removed, the characteristic neurovirulence pattern of Sabin Type II changed in a manner which can only be called a pattern of higher neurovirulence than NA-2, Type II, or original Type III, and the associated risk also became demonstrably higher than any of these. When Sabin Type III was cloned back to the neurovirulence pattern of Sabin original Type III, the associated risk dropped. It seems that monkey neurovirulence does have some relation to the risk to man.
Lot 56 was released on July 9, 1961. Lot 71 was released on January 31, 1962. There were no associated cases reported between those dates.
From January 31, 1962 until Pfizer vaccine cloned back to the neurovirulence pattern of the original Sabin Type III became available in 1964, there were 23 associated cases. One was excluded by D.B.S., leaving 22 as possibly vaccine induced. Of those, five were traced to Pfizer vaccine, one was either Pfizer or Lederle, and there were unknown. There were four Pfizer cases that could be pinned to one lot without alternativesthat lot was Lot 56. One case had three possible culpritsLot 56 was one of the three. No Pfizer lot but lot 56 was the solely implicated lot in any case after the release of Lot 56. Lot 56 could only be excluded as the implicated lot in 59% of the total cases reported, and the case now before the Court was not reported.
The government argues that D-100 establishes that Lot 56 had 20 million doses, and that the rate of vaccine association was lower than average, at most 1 in 4 million doses as compared with an average Type III risk one in 2½-3 million doses. However, the relevant statistic would be the number of doses of Lot 56 actually given, and this is very much a matter of conjecture. D-100 from its date and other factors, obviously represents an estimate of the number of doses in each total lot harvest. But this does not tell the whole story by any means.
The estimate for Lot 48 on the same sheet, D-100, is 2.51 million doses, but the filling records of Pfizer indicated that the final yield was only 2.03 million on packaging (D-65), a difference of around 17%. Further, the only evidence of field experience indicates that a 100 dose vial often did not yield 100 doses in the field, but up to 50% less (D-52). Vaccine had to be kept frozen and used within 1 year of packaging and seven days after unfreezing which would inevitably lead to more waste. To this must be added normal wastage from storage, transportation, breakage, etc. One reason given by Pitman Moore Company for not manufacturing live-oral polio vaccine was its estimate that, from its experience from similar products "the loss rate on the production of such a living viral product may approach 80% of the material harvested." (D-22).
The exact sale and distribution records as regards Lot 56 were, of course, not produced. It is also possible that Pfizer was selling from its stocks in Sandwich, England on the world market. The best that can be said is that, during the period in question, from licensing to the expiration of its lot life *23 in the Summer of 1964, probably somewhere between 4 and 15 million doses of Lot 56 were swallowed by human beings in the United States.
Lot 71 was even bigger than Lot 56. The protocol on Lot 71 (P-33) reveals that the size of the harvest was some 132 liters with a mean post-filtration titer of 7.9. Using the figure of 400,000 TCID50 per dose which was apparently used in figuring D-100, this gives approximately 29.7 million doses of Lot 71 at harvest. We know that 5 fillings of Lot 71 were recalled because of bacterial contamination. What percentage of the total of Lot 71 this represented is not shown. All in all, however, it would appear unreasonable to assume that less Lot 71 vaccine was administered in the U. S. during 1963 and the first half of 1964 than Lot 56 vaccine. Yet Lot 71 was never the sole lot implicated in a single case, and in the one case in which it was mentioned as a possible causal agent, Lot 56 was also given as an alternative.
While this does not establish that Lot 71 could not cause polio, (and in fact, if Mrs. Griffin did take a dose of Lot 71 in this case, Lot 71 did cause at least one case, hers) it seems more probable that the real culprit was Lot 56, even if there was a great deal more Lot 71 around during the Montgomery County campaign than Lot 56. Someone got the doses of Lot 56, and the natural probabilities springing from the relative quantities of 71 to 56 in a random case are not the most relevant consideration. If 100 people take pills, one of which is poison, the odds that any given person got the poison pill are only 99-1. If, however, one has the benefit of knowing that the subject in question died, the judgment becomes radically altered. Perhaps he died coincidentally of natural causes, or someone poisoned him before the test, or one of the other pills was unbeknownst to anyone poison. But most likely he got the poison pill.
Similarly in this case, there is some very remote possibility of other explanations, it is just possible that lot 71 actually induced this case, but the most reasonable conclusion is that Mrs. Griffin got a dose of Lot 56.
Mrs. Griffin took a dose of Pfizer Lot 56 polio vaccine, and was one of the tragically unfortunate few who contracted paralytic polio from it. But why should the United States be held liable for her truly horrible misfortune?
Prior to 1950, polio was a major crippler of children in this country, but little progress had been made in the search for a weapon to combat it. Around 1950, however, a number of discoveries paved the way toward the development of an effective vaccine, among them the fact that polio virus was an enterovirus, which grew first in the intestinal tract and spread afterward to the Central Nervous System through the blood, and the further fact that there were not a huge number of types of polio virus, as in the case of the common cold for instance, but rather only three types which were immunogenically homogeneous.
Many people set to work to develop vaccines. The first effective vaccine to come out of this research was Jonas Salk's killed virus vaccine. A killed virus vaccine is made by growing virus in a tissue culture and killing it clinically to render in incapable of causing disease, but in such a way that the dead virus particles are not chemically altered, and will still, when introduced into the body, *24 act as antigen and cause the body to produce an antibody which will, in the future, strangle any new live wild virus particles coming into the bloodstream, and therefore protect the central nervous system from infection and the inoculated person from disease. Salk developed a killed virus vaccine effective against each of the three polio types, and, after testing in the Frances Field trials, it was released for general use in 1955.
The Frances Field trials were a rigorous test of the safety and effectiveness of the Salk vaccine. They were carried out under the auspices of The National Foundation for Infantile Paralysis, popularly known as the March of Dimes. Detailed data were kept on all the test subjects in the trial.
The Frances Field trials demonstrated that properly killed Salk vaccine was both safe and effective. However, the dangers of improper killing procedure, and indeed, the need for strict and careful testing and surveillance when dealing with such potentially virulent agents as poliovirus, were graphically illustrated by what is known as the Cutter incident, which took place in late 1955. Somehow, some live virus survived the killing procedures in one Cutter Company distribution lot, and caused a number of cases of paralytic polio before the lot was recalled.
Although the Salk vaccine, properly prepared and administered, did and does provide complete protection against paralytic polio, it did not, during the years it was the primary weapon against polio, eradicate the disease in the United States. Salk vaccine had certain drawbacks which made it a reliable but rather slow tool in the war against polio. It was not the ideal vaccine for the ultimate in efficient immunization against polio, especially considering that polio struck hardest the lower socio-economic groups. First, Salk vaccine required hypodermic needle injection. Second, it required three separate injections. Third, it required booster shots every couple of years to remain effective. Fourth, it did not immunize the intestinal tract against infection by polio virus, so that persons immunized with Salk vaccine could still be links in the chain of infection of non-immunized persons. The Salk vaccine undoubtedly saved thousands upon thousands from the scourges of polio. However, many people felt that many thousands more could be saved if a more easily distributed cheaper vaccine could be developed, especially if that vaccine could provide intestinal immunity and also provide such permanent immunity that booster dosages were not necessary. Thus, in the late 1950's interest turned to the potentials of oral attenuated live-virus vaccine.
Dr. Albert Sabin had been working on such a live-virus vaccine approach for many years, and in the late 50's many others were also exploring this avenue, among them Dr. Herald Cox and Dr. Hilary Kaprowski.
In June of 1958, the then Surgeon General, Dr. Leroy E. Burney, realizing the great interest in live polio vaccine, and also that his office would be charged with the licensing and supervision of manufacture of such a vaccine when it was developed, appointed a committee of six experts on various aspects of the polio problem, hereinafter referred to as the "Surgeon General's Committee," or simply "the Committee," charged with keeping him abreast *25 of recent developments in the live-virus vaccine field and with recommending courses of action to be followed by the Surgeon General in promulgating regulations for the licensing and manufacture of live-virus vaccine.
The world community of medical science, including members of the Surgeon General's Committee, took part in the two initial conferences on the subject in June of 1959 and June of 1960.
At this time it would be well to explain the exact nature of the bureaucratic relationships which existed throughout the time pertinent to this litigation. At the top of the pyramid was the Secretary of Health, Education and Welfare. One of the divisions of the Department of Health, Education and Welfare was the Public Health Service, which was headed by the Surgeon General. The Public Health Service was divided into four divisions, one of which was called the National Institutes of Health, headed by a director who figures but slightly in the case at bar. One of the National Institutes of Health was called the Division of Biologic Standards (D.B.S.). The head of D.B.S. was Dr. Roderick Murray who was also Chairman of the Surgeon General's Committee by virtue of the fact that D.B.S. would be charged with the primary administration of any regulations promulgated by the Secretary of Health, Education and Welfare and the Surgeon General concerning the licensing of live polio virus vaccine. D.B.S.' in-house expert on testing live-virus preparations was Dr. Ruth Kirschstein. Although she was not an official member of the Surgeon General's Committee, she took active part in most discussions and was extremely influential on the final recommendations of the Committee.
The Committee reported directly to the Surgeon General, which is the reason the Director of the National Institutes of Health was not involved. Generally, any recommendations made to the Surgeon General by the Committee were accepted by the Surgeon General and Secretary of Health, Education and Welfare, and any recommendations on operational procedures and standards by the D.B.S. staff were accepted by the Committee. Further, after the adoption of the regulations governing the testing of production lots, D.B.S. was in primary charge of administering both testing and evaluation under the regulations, and held the power of approval or disapproval. The Surgeon General could have changed this power or overridden it. Sometimes the staff chose to consult the Committee. But the approvals and disapprovals of lots were issued by D.B.S. in the name of D.B.S. Further, it was Dr. Kirschstein who had primary responsibility for evaluating monkey neurovirulence tests. She signed the review approving Lot 56 for neurovirulence on July 5, 1961 (P-48) which led to its release on July 9, 1961 (D-96).
As early as August, 1959, the D.B.S. staff began to rough draft potential regulations for live polio virus strain selection and production (P-59, P-60) and submitted them to the Committee for review. Already three potential strains were beginning to appear the front runners in the race for first licensure, the Sabin strains, the Kaprowski strains, and the Cox strains, and of these the Sabin strains appeared to have an edge *26 over the competition in the minds of the Committee.
The evolution of D.B.S. thinking concerning the form of the regulations can be seen from a comparison of P-59 and P-60 of mid-1959, and P-63 and P-65 from mid-1960. By mid-1960 the rough regulations were beginning to approach their final form. The proposed standard for neurovirulence of production lots does not appear, but that adopted for strains was flatly that "No strain is considered acceptable unless its neurovirulence by each route of inoculation is no greater than that of the reference strain."
While D.B.S. was still adjusting the regulations, the Surgeon General's Committee met on August 19, 1960, with the D.B.S. staff and "other interested persons" and recommended that Sabin strains, Type I, II and III be selected for use in the U. S. (D-12). No regulations had yet been finalized or promulgated. The draft regulations had been changed only recently concerning the minimum criteria for strain selection (P-63 and 65). But the Committee recommended the acceptance of Sabin strains, without reference to the data upon which judgments of proper field experience were based. The Surgeon General ratified this by statements on August 24, 1960. Thus, it was found that the Sabin strains met requirements for strain selection which had not yet been finalized but which obviously were going to require some exact data on the field experience using adequate controls and surveillance, without reference to what data was considered or relied upon.
The proposed regulations were finally published in the Federal Register on November 23, 1961. They then contained exactly the same wording of the criterion for production lot neurovirulence evaluation as was finally adopted. After their publication, at least three influential persons suggested changes in that criterion toward the criterion representatives of D.B.S. at trial virtually admitted they used in actually judging the results of the monkey neurovirulence test. Dr. Stone suggested that the standard be made a demonstration that the production lot does not significantly exceed the reference strain, (D-28), Dr. Danielson suggested that it be demonstrated not to exceed "the range established for" the reference strain, (D-30), and Dr. Sabin, voicing his consistent criticism of the test assumptions of D. B.S. (which continued down to the day of his testimony at trial) suggested the inclusion of wording allowing passage of lots "in the same range." (D-27). Yet, the final regulation was promulgated as written on March 25, 1961. This Court therefore believes it was intended to mean what it said.
No vaccine which was not demonstrated to be no greater than NA-2 in neurovirulence was ever supposed to get to market. If an individual production lot *27 was not demonstrated by test to be so, it was not to be released.
The Government has urged that the decisions made by D.B.S. in this regard were discretionary. This is a misapprehension. (see discussion infra). They are not properly termed discretionary. They were duties created by valid regulation promulgated in accordance with the law. There was a positive duty upon the decision maker to apply the best standards available to the data at hand with all intellectual honesty in a professional manner to make the determination which the regulations required to be made. If there is room for difference among learned persons of the same profession, and if one of them is charged with the making of a decision and the other is not, then the fact that there is disagreement between the two will not render the decision of the individual charged with the responsibility of making the decision any less valid simply because there is room for doubt. But on the other hand, if the person charged with making the decision defines his postulates in such a way that no reasonable person could, with any intellectual honesty, reason from his postulates to the conclusion which he purports to make, he cannot hereby defend his conclusion by saying that there are other people with different postulates who could arrive at the opposite result.
By the time Lot 56 was released, D.B. S. knew that Type III was more genetically unstable than Types I or II, that Type III after the removal of Simian Virus 40 demonstrated a strange neurovirulence pattern marked by the presence of more severe lesions, had accepted the proposition that the intrathalamic test was the most discriminatory and most important test to be considered in determining neurovirulence, and knew that the test results of the laboratories of the drug manufacturers could not be trusted, such that they had to rely on their own test results in making judgments concerning the neurovirulence of production lots to be released. With those factors as given, and given the mandate of both the regulations and of any kind of common sense (and the principles of science are not at loggerheads with the principles of common sense in this instance), that the severity of lesions and the presence of paralysis in test monkeys be given reasonable weight in determining neurovirulence, this Court cannot imagine anyone saying that the test results available to D.B.S. "demonstrated that Lot 56 was of no greater neurovirulence then the reference strain." Quantitatively, Lot 56 had more lesions of all kinds than the mean experience with NA-2, and fell into the top of the distribution of the experience of NA-2 divided into 30 monkey lots. (See D-98) Qualitatively, the severity of the lesions demonstrated in Lot 56 are clearly in excess of anything ever encountered in NA-2. Further, Lot 56 demonstrated a paralyzed monkey, a phenomenon never encountered in NA-2. Given the fact that the intrathalamic test had already been accepted by D.B.S. as the most critical test, there is nothing in the results of the intraspinal test which would lead a reasonable individual to conclude in the face of the factors just enumerated that the test results demonstrated that Lot 56 did not exceed the reference strain in neurovirulence. It would be more accurate to say that the very opposite *28 was indicated. Therefore, Lot 56 should have been rejected.
Dr. Kirshstein testified that one of the reasons Lot 56 was passed instead of failed was that it was part of a series of other Pfizer lots that had demonstrated acceptable neurovirulence. The Court's response to that argument is that it is irrelevant. The lot to be tested was the lot in question. The reason for the test was to see if there was anything wrong with that lot. The fact that the four previous lots were good might say something for consistency of the manufacturing techniques utilized, but a lot can go wrong for reasons other than the consistency of manufacture, for instance, virulent mutation during in vitro passage. The monkey neurovirulence test was to test for these things as well as the harmful effects of bad manufacturing processes. The only regulation which said anything about comparing lots on the basis of previous lots, was one to test the consistency of manufacture (42 CFR 73.116(b)), which was in no way related to the evaluation mandated by 42 CFR 73.114(a) (1) (iii).
All of the evidence indicates that after it became obvious that the pattern of neurovirulence was different for Type III vaccine, D.B.S. began to evaluate Type III vaccine on the basis of quantity of lesions only. There is no real evidence, however, that D.B.S. ever honestly thought that the quantity of lesions was the only important factor in determining neurovirulence. It was, however, the only factor which could be used at the time to get any of the lots to pass.
The government witnesses argued that all the test results had to be viewed in the light of "biological variation." This is true as far as it goes, but it doesn't help the Government much. The concept of biological variation implies that, given a group of grossly homogeneous biological subjects (e. g. monkeys or humans), there will be or may be differences in response to a single stimulus. Most people don't get cancer from 20 years of smoking but some people do. The variations may be due to individual differences in the respondents, or manifestations of random factors in the interaction of the stimulus and the individual respondents (for instance, it has been argued that radiation-induced mutation may depend largely on how a given quantum of radiation strikes a particular cell). The argument of the government witnesses was that biological variation among the monkey test subjects would always lead to differences in test scores within a certain range among test groups of 30 even with a single homogeneous input such as the reference strain, which in fact was the case. Therefore, they argue, it was within the regulations to release any lot which fell into the range. There are a number of objections to this argument. First, biological variation works both ways some subjects will fail to respond to a high stimulus just as others will respond to a low stimulus. Partially because of this, given a known universe of experience, the results obtained are, unless more is known, more likely to be a median experience than any other specific experience. If you have a test on two monkeys and they show a result of X, this may be a skewed result. But, it is impossible without more to tell in what direction the skew occurs. If there were the entire universe of monkeys tested, the median result of those tested might show X-plus 2 or it might show X-minus 2 or X-minus 5 or X-plus 5 and without more the probability of variations the same above as below.
There was no more reason to assume the results of testing on Lot 56 was a high skew from a low median than that it was a low skew from an even higher median.
*29 Second, lot 56 was barely compatible with previous experience of NA-2 regarding incidence of lesions, and was not compatible with any rational statistical assumptions about experience with NA-2 for severity and spread.
The clear mandate of the regulation was that doubts be resolved against the lot under test. Dr. Kirschstein testified that Lot 56 seriously troubled and gave her pause. At that point, her only options under the regulations were further testing to resolve doubts or rejection. Yet she signed the review of testing approval which led to the lot's release.
Thus it is clear that at the time of the release of Pfizer Lot 56, the Lot 56 clearly did not meet the criteria of 73.114(b) (1) (iii) for release. The personnel of D.B.S. knew this or should have known it. It was released basically either because the personnel of D.B.S. charged with enforcing the regulation failed to read the regulation carefully, or because they did not understand or failed to take seriously the strict duty imposed upon them by the regulation. Either circumstance constitutes negligence.CONCLUSIONS OF LAW
1. This Court has jurisdiction over the parties and the subject matter.
*30 2. The place of the actions complained of was Bethesda, Maryland, the headquarters of D.B.S.
3. The Court must apply the law of Maryland to the facts of the case, including Maryland's choice of law rules.
4. The courts of Maryland would apply Pennsylvania substantive law to the facts at bar.
5. The regulation 73.114(b) (1) (iii) relating to neurovirulence testing in monkeys did not create a discretionary function under 28 U.S.C. § 2680(a).
6. The regulation created a duty to apply the scientific assumptions of the decision-maker to the facts at hand to reach an intellectually honest answer concerning the question which was made the operative question for release under the regulation.
7. That duty ran to the ultimate consumers of the vaccine tested.
8. The operative question created by regulation 73.114(b) (1) (iii) was whether a "comparative analysis of the test results demonstrate that the neurovirulence of the test virus pool does not exceed that of the NIH Reference Attenuated Poliovirus."
9. The U. S. Government, through its agents in the D.B.S. acting within the scope of their employment, breached that duty and negligently released Lot 56 for public consumption, in violation of regulation 73.114(b) (1) (iii), since a comparative analysis of the test results did not demonstrate that the neurovirulence of Lot 56 did not exceed that of the NIH Reference Attenuated Poliovirus."
10. Although recall of lots lawfully released originally which later became questionable through changed circumstance might be discretionary, there was a positive duty at all times to recall lots originally released negligently and in violation of regulation.
11. No new information was received by D.B.S. or any government agency from the negligent release of the lot to the time of Mary Jane Griffin's ingestion of a dose thereof which would vitiate that duty.
12. The United States, through its agents in D.B.S. was negligent in not recalling Lot 56 before October 22, 1963.
13. The negligent release and subsequent negligent failure to recall created a recognizable increased risk of harm to members of the class to whom the Government's duty was owed.
14. Mary Jane Griffin was a member of that class to whom the duty was owed in that she was an ultimate consumer of Lot 56 vaccine.
15. The injury to Mary Jane Griffin was foreseeable in that injury to some members of the protected class was foreseeable.
16. The right of Mary Jane Griffin to be free of the disease of Polio was protected against unintentional invasion.
17. The negligence of the United States was the proximate cause of Mary Jane Griffin's injuries.
18. Because of the injuries to Mary Jane Griffin, Richard Griffin was injured by the loss of her services as a wife.
19. The negligence of the United States was the proximate cause of Richard Griffin's injuries.
20. Neither plaintiff was guilty of contributory negligence.
21. The United States is liable to both plaintiffs for any damages caused by and through the negligence of the United States, to wit, the damages, stemming from Mary Jane Griffin's contraction of paralytic Poliomyelitis.DISCUSSION OF CONCLUSIONS OF LAW
It is best to deal first with the government's claim that this case falls within the discretionary action exception of the Federal Tort Claims Act, 28 U.S.C. § 2680(a). This exception has caused great problems of interpretation since its inception. The best judicial statement *31 of the factors to be considered in this case is found in Hendry v. United States, 418 F.2d 774, at 782 (2nd Cir. 1969), where the Court said:"No very clear rule emerges from either the licensing or the malpractice cases. It seems clear that where the grant of a license depends upon the balancing of several factors and the grant or refusal to grant is made without reliance upon any readily ascertainable rule or standard, the courts will hold the judgment to be discretionary. On the other hand, the Pennsylvania Railroad case, [Pennsylvania R.R. Co. v. United States, 124 F. Supp. 52 (D.N.J.1954)], supra, seems correctly to indicate that where the grant involves nothing more than the matching of facts against a clear rule or standard, the grant will be considered operational and not discretionary. Thus a hypothetical statute requiring that an employment certificate issue to all applicants of more than a given age, height, and weight would not empower an official to make a discretionary judgment under Section 2680(a)." "It is of course much easier to state this broad distinction than to apply it to the particular facts of a case. For the most part courts seem to have done so by their own intuitive judgment as to the difficulty of the administrative medical decision involved. Doubtless this proclivity explains why courts have found jurisdiction where medical mistakes have been clear, such as the use of the wrong medicine, but have declined to assert jurisdiction in the murkier areas of psychiatry. However, a distinction based solely on the difficulty of a decision would seem to confuse the existence of discretion under the statute with the substantive question of whether due care was lacking and how demonstrably it was lacking on a particular set of facts. Such an approach to decisions causes illogical results, as, for example, that the maintenance of physical security is operational in the case of a clearly suicidal patient, United States v. Gray, [199 F.2d 239 (10 Cir. 1952)], supra, but discretionary in the case of a patient whose aggressive tendencies were much less apparent. Dugan v. United States, [147 F. Supp. 674 (D. D.C.1956)], supra. Moreover, such an approach creates an exemption so broad that only in the most clear-cut instances of negligence can an injured person recover from the Government. To privilege from suit all governmental actions concerning which it conceivably might be demonstrated that there was a remote possibility that the officials arrived at correct judgments would be a severe limitation upon access to the courts in an age which has seen a steadily increasing impact by Government on the daily lives of its citizens." "We do not propose in place of this standard any `litmus paper test' as to whether discretion exists, Ove Gustavsson Contracting Co. v. Floete, 299 F.2d 655, 659 (2 Cir. 1962), cert. denied, 374 U.S. 827, 83 S. Ct. 1862, 10 L. Ed. 2d 1050 (1963). However, we do find a number of factors relevant in reaching a proper decision. First, it is pertinent to inquire whether the complaint attacks on the one hand the nature of rules which a government agency has formulated, or on the other hand the way in which these rules are applied. It is clear that § 2680(a) was intended to protect the validity of governmental regulations from challenge in a tort action for damages, Dalehite v. United States, 346 U.S. 15, 27, 73 S. Ct. 956, 97 L. Ed. 1427 (1953). The Dalehite Court stated, at 34, 73 S.Ct. at 967: `The "discretion" protected by the section is not that of the judgea power to decide within the limits of positive rules of law subject to judicial review. It is the discretion of the executive or the administrator to act according to one's judgment of the best course, a concept of substantial historical ancestry in American law. (Footnote omitted).'
*32 And at 35-36, 73 S.Ct. at 968:`* * * the "discretionary function or duty" that cannot form a basis for suit under the Tort Claims Act includes more than the initiation of programs and activities. It also includes determinations made by executives or administrators in establishing plans, specifications, or schedules of operations. (Footnote omitted).'" "This language appears to privilege from suit those decisions which either establish a rule for future governmental behavior or constitute an ad hoc determination which neither applies an existing rule nor establishes one for future cases. By contrast Dalehite seems to subject to suit those decisions which apply an existing rule to the facts of a case. To be sure, the application of existing rules in new contexts sometimes involves policy decisions by administrative officials akin to those made by legislators. Nevertheless, it can usually be determined whether the governing statute or regulation contemplates that an official will make new rules or ad hoc decisions on the one hand or apply old understood rules on the other. Whether the person whose judgment is attacked occupies a high or low level governmental position may be relevant in reaching a conclusion. Too, depending on whether it emphasizes the discretion of the decision-making officer or agency as did the regulation attacked in Dupree v. United States, [247 F.2d 819 (3 Cir. 1957)], supra, the language of the statute or regulation may guide the determination." "Another way to ask these same questions may be to inquire whether state law standards can adequately evaluate the course of action contemplated by a federal statute or regulation. This is not to imply that state tort standards, designed to remedy private wrongs, cannot apply to uniquely governmental decisions. See Indian Towing Co. v. United States, [350 U.S. 61, 76 S. Ct. 122, 100 L. Ed. 48], supra. However, state tort standards cannot adequately control those governmental decisions in which, to be effective, the decision-maker must look to considerations of public policy and not merely to established professional standards or to standards of general reasonableness. Of course the courts may inquire into whether an engineer's judgment in applying the rules of his craft was reasonable, but if the engineer is a government official who must, apart from questions of engineering efficiency and safety, determine whether a particular program of production is a desirable program, the courts cannot doubt the reasonableness of his evaluation of the public interest." "Here it seems clear that Hendry complains of the way in which the Coast Guard and the Public Health Service applied medical principles to his case, not of the content of the principles which they adopted as relevant. Nor did the doctors who conducted the examination of Hendry have policy-making rank in their organizations. Finally, the statute and regulations governing the delicensing procedure do not appear to convey discretion to identify and consider public safety goals. The only discretion apparently contemplated is that inherent in the judgments of any medical doctor in private practice." "Finally, complaints attacking discretionary decisions may frequently raise questions which are political and nonjusticiable in nature, but here the judgments arrived at by the doctors are not different in kind or complexity from those which courts are accustomed to entertain when tort suits are brought against private physicians. The fact that judgments of government officials occur in areas requiring professional expert evaluation does not necessarily remove those judgments from the examination of courts by classifying them as discretionary functions under the Act. To the extent *33 that the medical profession establishes no set rules to accommodate the handling of a particular medical case, the individual doctor's judgment in that case should be measured by the standards of due care."
In this case, the nature of the rules is not attacked, but rather the way the rules were applied. The history of the process of regulation-making in this case contains no implication that there was any intended delegation, should a new context arise, that D.B.S. should consider not only the outcome of applying the regulations honestly, but also the possible desirability of changing the regulations, sub rosa, to a standard of lower safety because of a judgment of net gain in the public welfare to be gained therefrom in the minds of the personnel charged with enforcing the regulations. Further, the fact that lots began to fail under the standard imposed by the regulation can hardly be called a "new circumstance." If the concept of safety regulation is to have any meaning, the possibility of failure, even widespread failure, must be taken as contemplated by the regulation. The discretion to revise a safety regulation in the direction of greater public risk should not be assumed absent clear language supporting such a conclusion in the regulation itself. There was none in this case. The public may not be subjected to the now-you-see-it, now-you-don't protection of safety regulations that a finding of discretion in this case would create.
The increased risk imposed upon the public by the disregarding of 73.114(b) (1) (iii) was statistically small. The regulations could properly have been promulgated in such a way as to impose that risk on the public had the initial judgment been that the benefits coming from such a course justified it. This was not the case, however. The regulations were promulgated by a procedure established to give hearing to any who might object to the balancing done by the agency, and once promulgated have the force of law and reflect the proper balancing between risk and benefit in the judgment of those lawfully empowered to make that judgment according to the process of the law. That judgment may only be changed by the process of the law. Any deviation from the regulations which imposes a higher risk on the public, no matter how slight, is not a matter of discretion. It may not be done.
The clear language of the regulation dictates that any doubt concerning whether the neurovirulence of the lot under test exceeds the reference must be resolved against the lot under test. This was a duty imposed by the promulgation of the regulation on those charged with administering it even if they had originally had a hand in drafting and promulgating it. The language is clear and without ambiguity or room for interpretation to anyone who would take the time to read it carefully. It is true that this was a judgment requiring professional expertise, but this does not render it discretionary. This Court is fully capable of scrutinizing the processes and conclusions of the decision-maker by the usual standards applied to cases of professional negligence. As explained at length, supra, the duty was clearly violated.
This Court has jurisdiction over both the parties and subject matter in this action under the Federal Tort Claims Act, 28 U.S.C. § 1346(b). The negligent acts complained of here took place in the State of Maryland, and therefore Maryland law must be applied, including Maryland's choice of law rules, Gowdy v. United States, 412 F.2d 525 (6th Cir. 1969) cert. den. 396 U.S. 960, 90 S. Ct. 437, 24 L. Ed. 2d 425, rehearing den. 396 U.S. 1063, 90 S. Ct. 750, 24 L. Ed. 2d 756. Under the conflicts rules of Maryland, where the lex loci delicto standard of Restatement I still prevails, a Maryland court would apply the substantive law of the State of Pennsylvania in a situation such as this, Uppgren v. Executive Aviation Services, 326 F. *34 Supp. 709 (D.Md.1971). This Court will apply the law of Pennsylvania to the facts. The Pennsylvania rule for negligent torts, articulated in Dahlstrom v. Shrum, 368 Pa. 423, 84 A.2d 289 (1951), is that of the Restatement, specifically that of Section 281(b) Restatement 2d, Torts, § 281, 1965, which states:"The actor is liable for an invasion of an interest of another if: (a) The interest invaded is protected against unintentional invasion, and (b) The conduct of the actor is negligent with respect to the other, or class of persons within which he is included, and (c) The actor's conduct is a legal cause of the invasion and (d) The other has not so conducted himself as to disable himself from bringing an action for such invasion.
Comment: See Risk to class of which plaintiff is a member. In order for the actor to be negligent with respect to the other, his conduct must create a recognizable risk to the other individually or to a class of persons as for example, all persons within a given area of danger of which the other is a member. If the actor's conduct creates such a recognizable risk of harm only to a particular class of persons, the fact that it in fact causes harm to a person of a different class to whom the actor could not reasonably have anticipated injury, does not make the actor liable to a person so injured."
In this case, it is obvious that the interest of Mary Jane Griffin in being free of poliomyelitis is protected against unintentional invasion by the law of Pennsylvania. Second, it is clear that the conduct of D.B.S. in releasing Lot 56 in violation of 42 C.F.R. 73.114(b) (1) (iii) was negligent per se by the law of Pennsylvania, Ennis v. Atkin, 354 Pa. 165, 47 A.2d 217 (1946). Third, it is clear that the negligence of the Government violated a duty owed to plaintiffs, in that plaintiff, Mary Jane Griffin was a member of the class of persons the regulations existed to protect, consumers of a particular lot tested, in this case, Pfizer Lot 56. See Anglo-American & Overseas Corp. v. United States, 144 F. Supp. 635 (S.D.N.Y., 1956) Aff'd. 242 F.2d 236 (2d Cir., 1957). Fourth, it is clear that the negligence of the United States was the proximate cause of plaintiff's injuries, because but for the negligence the harm would not have occurred, Coyne v. Pittsburgh Rys. Co., 393 Pa. 326, 141 A.2d 830 (1956); because the negligence set in motion forces which led to the injury which D.B.S. might or ought to have anticipated, even though in advance the injury seemed improbable and the precise form in which the injury resulted could not have been foreseen, DeFrangesco v. Overstreet, 8 Chest., 334, (Chester Co., Pa.1958); because the negligence created a significantly larger risk to the protected class of ultimate consumers which should have been foreseen and which came to fruition in the case of Mary Jane Griffin, and because the harm was the very injury intended to be prevented by the regulation, and must therefore be taken as proximately caused by its violation, Gatenby v. Altoona Aviation Corp., 268 F. Supp. 599 (W.D.Pa.1967), aff'd. 407 F.2d 443 (3rd Cir., 1968). It is also clear that neither plaintiff has done anything to disable himself or herself from bringing an action in this case.
Were D.B.S. a private corporation charged by the law of Pennsylvania with releasing vaccine lots to the public only in conformity with 42 C.F.R. 73.114(b) (1) (iii), liability would be clearly established under Pennsylvania law by the facts of this case. Since it is conceded that the employees of D.B.S. were acting within the scope of their employment when they released Lot 56, the United States is liable to plaintiffs under the Federal Tort Claims Act.DAMAGES
On October 27, 1963, Mrs. Griffin took a dose of Pfizer Lot 56 Type III polio *35 vaccine. On November 22nd she began to feel sick. At first she thought that she had some sort of influenza. Her symptoms progressed rapidly. By the afternoon she was suffering from a very very severe headache. By the next afternoon, the headaches had become so severe that a doctor was summoned. Her neck began to become stiff. The doctor recommended that she enter the hospital. She began to notice the first signs of paralysis in her legs and she was too sick to dress herself. Her knees wouldn't bend. She got into the car for the trip to the hospital and found that her spine wouldn't bend far enough to allow her to sit. When she arrived at the hospital, she found that she couldn't manage to climb onto the examining table by herself. She was given a sedative and her last conscious realization before it took effect was that she could no longer touch her face.
For the space of the next weeks, she was comatose. She hallucinated a great deal. Her occasional rises to consciousness came at such times as the performance of a tracheotomy and when a priest was giving her last rites. When she finally began to become conscious again, she found herself in an iron lung unable to speak, because of her open trachea, unable to move and being fed through tubes in her nose. There followed weeks in the iron lung at Lankenau Hospital and subsequently a transfer to Thomas Jefferson Hospital, where Mrs. Griffin was put first in a respirator and then as her rehabilitation slowly progressed, into a rocking bed, and finally after she had literally learned to breathe again, into a wheelchair. She also had to learn to control her bowels and her bladder again, and it was months before she obtained any control over these mechanisms.
After months of physical therapy, she was finally returned to her home. The present condition of her paralysis is as follows. She can move her right arm sufficiently at the elbow to touch her nose but not to touch the top of her head. She cannot brush or comb her hair, she cannot put makeup on the upper part of her face. She has almost normal strength in her right hand, but cannot use it for long periods of time without getting cramps. She can use her right hand to write if paper is placed in front of her, or to make telephone calls if the telephone is placed in front of her. Her left arm is completely useless. Her left shoulder is subluxated. It falls out of the socket. It causes excrutiating pain. Her arm has to be supported in order to alleviate the pain. Those who are taking care of her have to be extremely careful when dressing her so that the arm is not torn out of the socket. She has just enough use of the fingers of her left hand to be able to grasp a small object in a rather unsteady way, which allows her to open small objects such as compacts and lipsticks. She can lift the lower part of her right leg. She cannot lift her left leg at all, although she can contract the muscles in her things and calfs but without sufficient strength to move the limb. See can move her right leg sufficiently to kick her wheelchair back to occasionally change its position. Seating tolerance in an upright position is only two or three hours.
Two-thirds of her diaphragm is paralyzed and this creates a dangerous condition in that Mrs. Griffin, while able to breathe normally, is unable to use the diaphragm to cough or bring up phlegm and consequently mild respiratory infections always present the spectre of pneumonia. She currently has bowel and bladder control but no sphincter control so that her bowel and bladder movements must be carried out with the aid of another person on a time schedule. She has slight residual paralysis in her neck. Her life expectancy appears to be a normal 28 years. All the testimony was that there was nothing in her medical picture which would detract from her normal life expectancy since all of her internal organs were intact and unaffected except her diaphragm. While the *36 paralysis of her diaphragm does expose her to risks which are greater than that for the average individual, these risks are minimized by the constant attention that she receives from her husband and her nurses. The quality of this attention has been so high and so carefully constructed in the past, that the dangers brought about by her condition do not seem significantly to detract from her normal life expectancy of 28 years.
The hospital and medical bills to the time of trial totaled $89,223.25. This is based on a bill of $2,209.85 for her hospitalization at Lankenau Hospital, a bill of $5,712.00 for her stay at Jefferson Hospital, a bill of $11,645.00 for her nursing care during those hospital stays, a bill of $723.40 for hospitalizations between 1963 and the time of trial resulting from respiratory infections brought about by virtue of Mrs. Griffin's paralyzed diaphragm, $7,067.00 in medicine and drugs, $52,727.00 for nursing expenses from 1964 through 1971, purchase and maintenance of $2,491.00 worth of medical equipment which was necessary for the continued health and physical rehabilitation of Mrs. Griffin, and $6,648.00 which is based upon the cost of changes to the house brought about by medical necessity in order to insure that Mrs. Griffin can continue living, the total cost of which were $8,648.00 but which resulted in an added value to the house of around $2,000.00. These expenses were fair, necessary and reasonable to the treatment of her injuries and to the maintenance of her life function in the face of her injuries.
As to future medical expenses, the Court has examined all the calculations of plaintiff's expert, Dr. Verzilli, in arriving at what the plaintiffs have requested as an award for future medical expenses and has concluded that virtually every assumption that Dr. Verzilli made was a conservative one. Dr. Verzilli based his suggestion not on an inflationary rise in prices, but on a rise in prices attributable to the average rate of economic growth of 2½% per year, which was in itself a rather conservative estimate. All figures were rounded off low, and the projection was based on a 25 year expected life. In view of this, plaintiffs' request for an award of $713,354.00 in future medical expenses reduced to present worth to $421,581.00 in future medical expenses, is entirely reasonable and will be granted. As to future earning capacity, it seems reasonable to assume considering both Mrs. Griffin's personality and propensity to work, that she would have, had she not been paralyzed, returned to full time gainful employment at something around the date of trial considering the ages of her two daughters at the time of trial. Again, the Court has examined Dr. Verzilli's calculations in his testimony at trial and again the Court finds that his assumptions in making his calculations were quite conservative, assuming a 3% a year growth and assigning Mrs. Griffin only the median income for working women in this area, when it appears to the Court that a woman of Mrs. Griffin's experience and talents would probably have been able to make much more. With these factors in mind, it appears reasonable to find that Mrs. Griffin's future earning capacity would have been at least $89,000.00, which reduced to its present worth, comes to $61,428.00. Her earning capacity through her telephone appears from the evidence to be about 20% of what it otherwise would have been. Reducing the above figure accordingly, yields an award for impairment of future earning capacity of $49,142.40.
As to pain and suffering and the ability to enjoy life's pleasures before as after the accident, this Court notes that Mary Jane Griffin cannot leave her home except on rare occasions and that she cannot enjoy, obviously, any of life's pleasures, requiring physical exertion from any of her paralyzed limbs. She suffers excruciating physical pain from time to time and is always in danger of being subjected to that pain. She undergoes the kind of mental *37 suffering that only a quadriplegic who had lived an active life before her paralysis can know. She is completely aware, she is completely alert, she is sensitive to every nuance of both physical and mental anguish. She has become completely dependent on other persons, even to her bowel and bladder functions. She has become her husband's jailer. She knows these things. She can look forward to being nothing but these things. The law says that she must be compensated in money for her damages, for the law knows only to compensate in money for damages which are really beyond value.
To begin this task in this case, this Court has turned to recent decisions of other courts dealing with catastrophic injuries for guidance, to wit, Frankel v. United States, 321 F. Supp. 1331 (E.D. Pa., 1970) aff'd. 466 F.2d 1226 (3rd Cir. 1972), Schwartz v. United States, 230 F. Supp. 536 (E.D.Pa.1964), Christopher v. United States, 237 F. Supp. 787 (E.D. Pa.1965) and Tinnerholm v. Parke Davis & Co., 285 F. Supp. 432 (S.D.N.Y.1968). But all of these cases present special considerations which, to this Court's mind, render them, if such a judgment may ever be valid, less catastrophic than the present case. In Frankel, there was some question as to the quality of awareness of the plaintiff and there was the possibility that money awarded would work less as a compensation for her than as a windfall to others. The period of time involved was significantly less. In Schwartz, the disfigurement was truly horrible, yet the plaintiff was in no way reduced to complete physical dependency on others. In Christopher, the plaintiff was only paralyzed from the waist down, a terrible condition, but vastly preferable to that of the present plaintiff. In Tinnerholm, the paralysis was only partial, and there was again great question as to the quality of awareness and the problems of an award acting as a windfall to third parties rather than any actual compensation to the injured party.
There is certainly room for reasonable men to differ on the exact size of the award called for by injuries such as plaintiff's. Yet the decision must be made. In the final analysis, the Court is charged with the lonely agony of finding a common ground between what it knows from its experience of the value of money, and the value of the lost pleasures and the agony of a particular person. It is undoubtedly true that Mrs. Griffin's physical pain was more exquisite in the first year of her illness. Yet, her mental anguish is a daily condition.
Physical pain which will and can often be borne more easily than mental pain which will not. Further, the obvious inability to enjoy life's pleasures after the injury as before, is everpresent. The Court does not really believe that any rational purpose would be served by valuing Mrs. Griffin's pain and suffering of any one time greater than that of any other. The Court, therefore, makes a lump sum award in compensation for Mrs. Griffin's pain and suffering of the nine (9) years past and the twenty-eight (28) years future, on consideration of the particular severity of her condition, coupled with her unaffected alertness and capacity to experience her pain and loss as everpresent and continuing agony, of One Million Two Hundred Thousand ($1,200,000.00) Dollars.
As to Richard Griffin's claim for loss of consortium, past and future, the law says that he is to be compensated for the economic value of his wife's services and also for those services which cannot be valued. It is obvious that Mr. Griffin has lost the total economic value of his wife's services. She has gone from being a financial asset to a financial liability.
In many cases there is an attempt to prove this economic value by dividing a wife up functionally and showing how much it would cost to hire someone to perform each function. The results are astronomical, but artificial. Yet who, in *38 this age of woman's liberation, could deny that a wife's tangible labor is worth as much as that of the average factory worker, which was $7,462.52 per year in 1971. The economic value of the 29 lost years of Mary Jane Griffin's services at this rate would be $216,413.08. Now the Court is aware that there would be many adjustments (for economic growth, reduction to present worth of the future portion, etc.) which would have to be made for this figure to make economic sense as an award by itself. But it gives an idea of the magnitude of the purely economic value of the labor of a wife.
An economic analysis can only be another impressionistic jumping off point for an evaluation of the damage done to a husband for loss of consortium. In the present case, given the continuing nature of the losses, the improbability of any mitigation of that continuing nature to the end of the husband's life, and the evidence of the value to the husband of this particular relationship as a whole which comes from his actions in fulfilling his responsibilities to his wife in her current condition, this Court finds that an award of $300,000.00 for past and future loss of consortium would be reasonable and just.
Accordingly, this Court finds for Mary Jane Griffin and against the United States in the amount of $1,759,946.25, and for Richard Griffin and against the United States in the amount of $300,000.00. The issue of the effect of the Joint Tortfeasor's release between plaintiffs and Pfizer on the amount actually due plaintiffs will be ruled upon after further briefing by the parties.NOTES
 The discussion of the evidence draws on certain circumstantial facts which technically do not hare to be "found", (although they may be), as they are not outcome determinative. Insofar as the Court is exercising its prerogative as finder of fact to determine the weight and credibility of the evidence, however, the discussion may also be regarded as containing supplementary findings of fact.
 Through the remainder of this discussion, plaintiffs' exhibits are designated simply by a "P" followed by a dash and a number, and defendant's exhibits are designated simply by a "D" followed by a dash and a number.
 The information in the letter from C. F. Hagan to the Justice Department in 1972 (D-93) is obviously taken from D-92.
 It is perhaps technically improper to speak of a burden of persuasion as to such a circumstantial fact. A burden of persuasion is imposed by a mechanism properly called an assumption, by which the law, out of necessity, deems a necessary fact untrue until it is proven to a given degree of certainty (the standard of proof; in this case, a preponderance of the evidence). See generally, Ashford & Risinger, Assumptions, Presumptions & Due Process in Criminal Cases; A Theoretical Overview, 79 Yale L.J. 165 (1969). As to a circumstantial fact, the law needs no rule of decision to deem it either true or untrue, but simply lets the evidence in doubtful cases stand on its own inferential feet as it is taken with other circumstantial factors into the calculus which finally yields a decision from the trier of fact on the ultimate necessary question on which it bears.
 Another explanation, also speculative, would run as follows: B73088 might be a superceding invoice representing the end result of a previous shipment and credit, reflected by the pencilled addition on its face. There are probably other such speculative explanations which could be constructed.
 It isn't really very credible. All type III feedings in the mass campaign were over before November 19 (the type III makeup feeding was November 3(P-1) and the last institutional feeding shown was November 7(P-21).
 See the detailed record keeping requirements of 42 CFR 73.37 at footnote 8 infra.
 42 CFR 73.37 a & b, promulgated July 28, 1961, requires:
(a) Maintenance of records. Records shall be made, concurrently with the performance, of each step in the manufacture and distribution of products, in such a manner that at any time successive steps in the manufacture and distribution of any lot may be traced by an inspector. Such records shall be legible and indelible, shall identify the person immediately responsible, shall include dates of the various steps, and be as detailed as necessary for clear understanding of each step by one experienced in the manufacture of products.
(b) Records retention(1) General. Records shall be retained for such interval beyond the expiration date as is necessary for the individual product, to permit the return of any clinical report of unfavorable reactions. The retention period shall be no less than five years after the records of manufacture have been completed or six months after the latest expiration date for the individual product, whichever represents a later date.
The records of manufacture were not complete until D.B.S. released Lot 56 on July 9, 1962. Thus Pfizer had a duty by regulation to retain its detailed lot distribution records until July of 1967, at which time it had already been sued by the Griffins.
 See footnote 18 infra.
 P-39, Committee report, p. 4.
 Except for one case of death possibly attributable to Coxsackie A-7 reported from Scotland which defendants' expert Dr. Fox rather archly implied was a form of permanent quadriplegic paralysis.
 The only blood test findings on antibody titers in this case were, through no fault of anyone, done at a point in time so remote as to render them of little value.
 The evidence adduced at trial does not reveal the exact release date of Pfizer lots 93-97, which were the cloned lots, but it appears to have been no later than the summer or fall of 1964. They were in preparation in early 1963 (P-55) and finally tested by D.B.S. in November of 1963 (P-47).
 Estimates are complicated by the sale of some 7 million doses of Type III vaccine by Pfizer to Wyeth in September 1963, which were then distributed under Wyeth's label; thus the 4 cases attributed to Wyeth after that date might really be traceable to Pfizer vaccine. It was not shown whether lot 56 vaccine was or was not sold to Wyeth.
 The titer represents the base ten exponent of the number of TCID50 per milliliter. Multiplying 107.9 times 1000 gives the TCID50 per liter, and multiplying this times the number of liters in the lot harvest gives the TCID50 in the harvest.
 As noted above, D-100 is too early to represent actual filling information. Yet the average TCID50 per dose yielded by the titer and dosage figures given there varies a bit. This may have arisen from figuring the lot sizes at different times with different average dose assumptions. More likely the titer figures were originally two significant digits, and were used that way to figure the TCID50 volume, then rounded off for the chart.
 This Committee consisted, at all times pertinent hereto, of Dr. Roderick Murray, Dr. David Bodian, of John Hopkins University; Dr. William McD. Hammon, of the University of Pittsburgh School of Public Health; Dr. Alexander Langmuir, of the Public Health Service's Communicable Disease Center, Atlanta, Georgia; Dr. Joseph Melnick, of Baylor University, and Dr. John R. Paul, of the Yale University Medical School.
This Committee has gone by a number of names. Originally it was called "the ad hoc Committee" (D-1), then the "Public Health Service ad hoc Committee on Live Polio virus Vaccine." (D-2), then "the Public Health Service Technical Advisory Committee on Live Polio virus Vaccine" (D-4), then "The Public Health Service Committee on Live Poliovirus Vaccine" (D-5), then simply the "Committee on Live Poliovirus Vaccine" (D-7), then the Surgeon General's Committee on Live Poliovirus Vaccine," (D-67).
The reason that this has been confusing is that the special 1962 Committee to investigate vaccine related cases also went by six different names at different times, similar to each other and to the names given to the Surgeon General's Committee set out above.
 The regulations pertinent to this action were promulgated by the joint authority of the Secretary of Health, Education and Welfare and the Surgeon General under 42 U.S.C. § 216, a procedure subsequently abolished by Reorganization Plan #3 of 1966 (Appended as a footnote to 42 U.S.C.A. § 202).
 This act was properly discretionary, since no regulation then in effect foreclosed such a judgment. The regulations as finally adopted would probably have foreclosed that judgment as it was made, however. No one called upon to do so at trial could document the tests in 100,000 susceptibles with adequate surveillance required by 42 CFR 73.110(b) (II) (i) as adopted, although each testified that it was his impression that the Sabin strains had to meet the standards of the regulations even though they were approved some 7 months before the final promulgation of the regulations.
 Actually, the evidence indicates that the real criterion utilized to pass lot 56 and other questionable lots was whether the total number of monkeys showing lesions of all kinds was within the range of previous experience with NA-2, regardless of the severity of lesions. The claim that tests were valuated both quantitatively and qualitatively to see if they fell within the possible expected range of NA-2 does not appear to be true. However, it is the closest anybody from the government ever got to implying that the standard created by the regulations was followed. Even this is in violation of 73.114 (b) (1) (iii) however.
 See footnote 23 infra.
 The reliance by the government on the testimony of Dr. Albert Sabin is a perfect example. Dr. Sabin has been consistent from the beginning of the discussion of possible regulation in 1960 and before in claiming that the intraspinal test was the most important test of monkey neurovirulence. This contention was repeatedly rejected by the D.B.S. staff. Dr. Sabin believes that the regulations as drawn were much too stringent. It is interesting to note, however, that even by Dr. Sabin's own suggested changes to the proposed regulations made in December of 1960, the presence of a paralyzed monkey would have required further testing before the lot could be released. (D-27, p. 3).
 The term "demonstrate" implies a fairly stringent standard of proof in the empirical sciences.
 It might be argued that at least here the experience of D.B.S. with previous lots provides some evidence of a high skew from a low median. But such an evidentiary assumption would undermine the integrity of the very concept of lot testing, have little effect on judgments based on grade 4 lesion severity and paralysis, where there was no previous experience with NA-2, and be very shaky in view of type III's known instability and the extremely high neurovirulence of the first 5 rejected Pfizer licensing lots.
 Dr. Maloney's development of acceptance tables for Type III lots based on statistical analysis are interesting. They are based on the Chi square test, which is a perfectly good statistical test mathematically. Unfortunately, as in the computer sciences, what comes out depends on what goes in, and the results Dr. Maloney obtained are a perfect example of the old computer programmer's chestnut "garbage in, garbage out." His January 1972 table is based on the rather remarkable assumption that D.B.S. experience to that date with NA-2 represented only a .05 probabilitythough he seems to know that an assumption of .50 is most rational. However, he also must have known that few Type III lots would pass on that basis without an outrageous consumer risk assumptionand didn't even bother to work up the figures. His June 1963 table (which was enthusiastically adopted by both the Committee and D.B.S.) abandoned all pretense of referring to NA-2 for judgments of acceptable lesion severity levels. Experience with NA-2 had become so great that virtually no assumption could make most Type III lots pass, so be based his table on the experience with Sabin original Type III, in complete violation of the regulations. These tables are perfect examples of the D.B.S.Surgeon General's Committee propensity to blithely violate the regulations to get Type III lots to pass rather than working to change the regulations while doing their duty by failing the lots. Lot 56 would have been rejected by both charts, incidentally.
 Notes of testimony, p. 870.
 Apropos of this, the following exchange took place at trial between Mr. Adler and Dr. Kirschstein:
Q (By Mr. Adler) Doctor, lets talk about biological variation. Humans are not all the same, just as monkeys are not all the same?
A (By Dr. Kirschstein) That is correct.
Q And therefore when you let loose this vaccine into the community of people, you have to be concerned that the humans are going to have a biological variation?
A That is correct.
Q And the NA-2 regulation didn't say "within biological variation," it said, "Not to exceed," didn't it?
A I don't believe it is stated that way. I believe it is stated as "being comparable to."
Q Well, let me read it to you.
Q It says:
"The virus pool under test is satisfactory for polio virus vaccine manufacture only if at least 80 percent of the animals in each group survives the observation period and if a comparative analysis of the test results demonstrate that the neurovirulence of the test virus pool does not exceed that of the NIH Reference Attenuated Polio Virus."
A I am sorry.
Q It doesn't say "comparable."
A Excuse me.
Q And you were looking for comparable results, weren't you?
 See Manufacturing Production Worker Statistics (U.S. Dept. of Labor), 1972 World Almanac, p. 408. These are the latest figures now available to the Court, and are sufficient for the Court's purposes in illustrating the general magnitude of the purely economic value of a wife's services.
 This 29 years reflects the nine years past and Mr. Griffin's future life expectancy of twenty years.