Richard J. Griffin and Mary Jane Griffin, His Wife v. United States

OPINION OF THE COURT

ROSENN, Circuit Judge.

This appeal raises important questions concerning the Government’s liability under the Federal Tort Claims Act, 28 U.S.C. §§ 1346(b), 2671 et seq.¹ Mary Jane Griffin and her husband, Richard Griffin, seek damages from the United States for injuries sustained by Mrs. Griffin allegedly as a result of ingestion of Sabin oral live-virus polio vaccine. • The district court, after a two week non-jury trial, rendered a judgment in favor of the Griffins and awarded damages in the amount of $2,059,946.25. The Government has appealed challenging:

(1) The district court’s failure to hold the action barred because the claim is based upon the exercise or performance of a “discretionary function,” 28 U.S.C. § 2680(a) ;²

(2) The district court’s findings on negligence and proximate cause;

(3) The district court’s award of allegedly excessive damages; and

(4) The district court’s failure to give effect to a joint tortfeasor release given by the Griffins to Charles Pfizer & Co., the manufacturer of the vaccine Mrs. Griffin ingested.³ We affirm the judgment of the district court as to liability and computation of damages but reverse its failure to give effect to the joint tortfeasor release.

In the fall of 1963, Mrs. Griffin participated in a program sponsored by the Montgomery County Medical Society to secure immunization against polio through the mass administration of the Sabin oral polio vaccine. The program called for three separate doses of vaccine to be administered on three different dates. Each dose was designed to protect against one of three particular polio virus types. On September 22, 1963, Mrs. Griffin ingested a dose of Sabin Type I vaccine at the Gladwyne School in Montgomery County, Pennsylvania. On October 27, 1963, she ingested a dose of Sabin Type III vaccine.^{4 5}On November 22 she began to feel sick. By the following day her symptoms progressed to the point where hospitalization was required. She tragically emerged from the hospital one month later a permanent quadriplegic.

In 1965 the Griffins commenced four separate actions. Two were instituted in the Court of Common Pleas of Montgomery County against the Montgomery County Medical Society. Two were instituted in federal court — one against Charles Pfizer & Co., and the second, the instant action, against the United States under the Federal Tort Claims Act. In 1971, the federal court action against Pfizer was settled for $350,000 and the Griffins executed a joint tortfeasor release in favor of Pfizer.® The plaintiffs obtained final judgment in the instant action on January 18, 1973.

In a thoughtful and thorough opinion, Judge Newcomer found that as a result of ingesting the Type III vaccine on October 27 Mrs. Griffin developed polio. The court found that the dose ingested by Mrs. Griffin was part of Pfizer production Lot 56. The court further found that Lot 56 had been subjected to testing for safety and potency by the Division of Biologic Standards (DBS), a division of the Department of Health, Education and Welfare.^{6 7}On the basis of undisputed test results, the court found that Lot 56 was approved for release to the public by DBS in violation of agency regulations. Specifically, the court held that the release of Lot 56 was inconsistent with 42 C.F.R. § 73.114(b) (1) (iii) in that:

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 neuro-virulence.

351 F.Supp. at 16. The court concluded that Mrs. Griffin was a member of the class of persons the regulation was designed to protect, and that the hazard and particular harm she suffered were those the regulation was designed to prevent. Accordingly, the court found that the approval of Lot 56 by the officials of DBS constituted negligence per se. The court held 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.” 351 F.Supp. at 34. Stating that “ [i] n this case, the nature of the rules is not attacked, but rather the way the rules were applied,” the court held that this case did not fall within the discretionary function exception to the Federal Tort Claims Act.

The court awarded damages for Mrs. Griffin’s past and future medical expenses, future earning capacity, and pain and suffering in the amount of $1,759,946.25. Mr. Griffin was also awarded $300,000 for past and future loss of consortium. The court rejected the Government’s argument that the plaintiffs’ recovery should be reduced by 50 percent due to the terms of the joint tortfeasor release. This appeal followed.

I. Discretionary Function

The threshold question confronting us is whether this action is barred because of the “discretionary function” exception ⁸ to the Torts Claims Act. Although there is some dispute as to whether the exception is jurisdictional or merely a defense available to the Government, this circuit has treated the exception as jurisdictional. Gibson v. United States, 457 F.2d 1391, 1392 n.1 (3d Cir. 1972).

The Government contends that the decision to release Lot 56 involved the exercise of a discretionary function. It argues that the determination called for by the regulation [42 C.F.R. § 73.-114(b) (1)(iii), supra note 8] that the neurovirulence of a particular lot does not exceed that of the “reference strain” involves the exercise of judgment. It maintains that Congress intended, by the discretionary function exception, § 2680(a), to exclude all claims “arising from acts of a regulatory nature.”

We believe that the construction of § 2680(a) urged upon us by the Government is too broad. Activity of any consequence is rarely without its judgmental component. The effect of accepting the Government’s contention would effectively immunize all Governmental activity from judicial review except the most ministerial acts. In its landmark decision, Dalehite v. United States, 346 U.S. 15, 73 S.Ct. 956, 97 L.Ed. 1427 (1953), the Supreme Court explicitly recognized that not all activity involving judgment is necessarily encompassed within the Act’s exception :

The “discretion” protected by the section is not that of the judge — a 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.

346 U.S. at 34, 73 S.Ct. at 967 [footnotes omitted]. The decisions held discretionary in Dalehite involved, at minimum, some consideration as to the feasibility or practicability of Government programs. 346 U.S. at 41, 73 S.Ct. 956. Such decisions involved considerations of public policy, calling for a balance of such factors as cost of Government programs against the potential benefit. The Court stated;

[The discretionary function] also includes determinations made by executives or administrators in establishing plans, specifications or schedules of operations. Where there is room for policy judgment and decision, there is discretion. [Emphasis supplied.]

346 U.S. at 35, 36, 73 S.Ct. at 968 [footnotes omitted]. Where decisions have not involved policy judgments as to the public interest, the courts have not held the decisions to be immune from judicial review. Eastern Air Lines v. Union Trust Co., 95 U.S.App.D.C. 189, 221 F.2d 62 (1955), aff’d 350 U.S. 907, 76 S. Ct. 192, 100 L.Ed. 796 (1955); Hendry v. United States, 418 F.2d 774 (2d Cir. 1969). See also Ward v. United States, 471 F.2d 667 (3d Cir. 1973); Pigott v. United States, 451 F.2d 574 (5th Cir. 1971). To determine the applicability of the discretionary function exception, therefore, we must analyze not merely whether judgment was exercised but also whether the nature of the judgment called for policy considerations.

At the outset, we emphasize what is not being challenged on this appeal. Plaintiffs do not challenge the Surgeon General's determination to approve a live-virus immunization program.⁹ Neither do plaintiffs challenge the regulation which established the standard against which all manufactured lots were to be measured. See note 8 supra. These were matters involving balancing of policy considerations in advancing the public interest. Plaintiffs, in the instant case, challenge solely the manner by which the regulation was implemented. They contend that in approving a particular lot, Lot 56, for release to the public, DBS failed to comply with the standard established by the Surgeon General.

The issue before us, therefore, is whether the implementation of regulation 73.114(b) (1) (iii) by DBS involved a “discretionary function.” To decide this question we must first determine exactly what the regulation required be done in determining whether to release a particular lot.

The crucial action in approving a particular test lot for polio vaccine manufacture was the determination that the neurovirulence¹⁰ of the test lot “[did] not exceed” that of the NIH “reference strain.”¹¹ The regulation required a “comparative analysis” of the monkey neurovirulenee test results of a particular test lot with the monkey neuroviru-lence test results of the reference strain.¹² The regulation enumerates five criteria as evidence of neuroviru-lence: the number of animals showing lesions characteristic of poliovirus infection, the number of animals showing lesions other than those characteristic of poliovirus infection, the severity of the lesions, the degree of dissemination of the lesions, and the rate of occurrence of paralysis not attributable to the mechanical injury resulting from inoculation trauma.¹³

Plaintiffs contend that the test lot could not be approved if it exceeded the reference strain with respect to any one of the five enumerated criteria. Under this interpretation of the regulation DBS could not approve a lot which minimally exceeded the reference strain with respect to any one criterion, even if DBS considered that criterion the poorest indicia of neurovirulenee of the enumerated criteria, and even though the test lot was far superior to the reference strain with respect to the other four criteria.

We do not agree with this construction of the regulation. The regulation merely lists five criteria as evidence of neurovirulenee and calls for a “comparative analysis.” DBS has consistently construed the regulation as permitting it to weight the criteria in accordance with the degree to which it believed each criterion reflected neurovirulenee.¹⁴

The Supreme Court has stated on another occasion:

Since this involves an interpretation of an administrative regulation a court must necessarily look to the administrative construction of the regulation if the meaning of the words used is in doubt .... [T]he ultimate criterion is the administrative interpretation, which becomes of controlling weight unless it is plainly erroneous or inconsistent with the regulation.

Bowles v. Seminole Rock and Sand Co., 325 U.S. 410, 413-414, 65 S.Ct. 1215, 1217, 89 L.Ed. 1700 (1945). We find the DBS interpretation of the regulation to allow weighting of the five criteria of neurovirulenee neither “plainly erroneous” nor “inconsistent with the regulation.” ¹⁵

We acknowledge that under DBS’ construction of the regulation, the implementation called for a judgmental determination as to the degree to which each of the enumerated criteria indicated neurovirulenee in monkeys. The judgment, however, was that of a professional measuring neurovirulenee. It was not that of a policy-maker promulgating regulations by balancing competing policy considerations in determining the public interest. Neither was it a policy planning decision nor a determination of the feasibility or practicability of a government program. At issue was a scientific, but not policy-making, determination as to whether each of the criteria listed in the regulation was met and the extent to which each such factor accurately indicated neurovirulenee.¹⁶ DBS’ responsibility was limited to merely executing the policy judgments of the Surgeon General.^16A It had no authority to formulate new policy in the immunization program.

Where the conduct of Government employees in implementing agency regulations requires only performance of scientific evaluation and not the formulation of policy, we do not believe that the conduct is immunized from judicial review as a “discretionary function.” As Judge Waterman of the Second Circuit has stated:

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.

Hendry v. United States, 418 F.2d 774, 783 (2d Cir. 1969).¹⁷ As the district court stated:

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.

351 F.Supp. at 33.

Even were we to conclude arguendo, however, that DBS’ approval of vaccine lots for release, had it complied with the regulation, was a “discretionary function,” we would not hold plaintiffs’ action barred by the “discretionary function” exception in the instant case. The Government’s release of Lot 56 was predicated upon its reliance on a factor called “biological variation.” Reliance on this factor, however, was not authorized by the regulations. We therefore conclude, as discussed below, that DBS’ activity was not immunized from judicial review.

The district court found that the undisputed test results showed that Lot 56 “probably exceeded” the reference strain in neurovirulence.¹⁸ DBS also apparently believed that if only the raw test results were considered the neurovirulence of the lot was greater than that of the reference. In ultimately approving Lot 56, however, the testimony showed that DBS did not consider the test results to be a conclusive indication of neuroviru-lence. Rather, the testimony indicated that Lot 56 was approved, despite the unfavorable test results, because the DBS officials considered the test results of that lot to be within “biological variation.” ¹⁹ The concept of “biological variation” is premised on the fact that a group of similar subjects will respond differently to a single stimulus. The variation in response is due not to variations in the stimulus but rather to differences in the subjects. As applied to the neurovirulence tests, DBS reasoned that the unfavorable test results of Lot 56 were not produced by defects in the vaccine sample tested, but rather were a product of unusual characteristics of the particular monkeys inoculated with the lot. The effect of relying on “biological variation” was to discount, or regard as not significant, what otherwise appeared to be excessive neurovirulence.

We are unaware of any authority conferred by the regulation to permit DBS to discount the results of particular tests because of “biological variation.” 42 C.F.R. § 73.114(b) (1) (iii) provided that

only if . the test results demonstrate that the neurovirulence of the test virus pool does not exceed that of the NIH Reference Attenuated Poliovirus

is the test lot satisfactory for vaccine manufacture. On its face the regulation required that DBS consider all test results as a meaningful indication of neurovirulence.²⁰ Moreover, a review of the comments and criticisms submitted to the Surgeon General in response to the notice of proposed rulemaking compels the conclusion that the results of testing were not to be diluted on this basis. The suggestions of several scientists that this language be modified to permit a more flexible standard, including “biological variation,” ²¹ were apparently rejected. The regulation, both as proposed and as promulgated, contained the strict requirement that the test results demonstrate that the neuroviru-lence of the test lot not exceed that of the reference strain.

Even were we to concede that discretion was otherwise conferred upon DBS by the regulation, no discretion was conferred to disregard the mandatory regulatory command. In discounting test results that were required to be considered significant, DBS acted outside the scope of the authority conferred by the regulation. The violation of a non-discretionary command takes what otherwise might be characterized as a “discretionary function” outside the scope of the statutory exception.²² United Air Lines v. Wiener, 335 F.2d 379, 393-394 (9th Cir. 1964), cert. dismissed, 379 U.S. 951, 85 S.Ct. 452, 13 L.Ed.2d 549 (1964) (Air Force Command’s designation of permissible flying areas, although an unreviewable “discretionary” determination had regulations been complied with, held removed from § 2680(a) exception by Command’s failure to make prior study required by regulation); see Hatahley v. United States, 351 U.S. 173, 76 S.Ct. 745, 100 L.Ed. 1065 (1956). Although Congress has provided that activity within the authority of a Government official, if a “discretionary function,” may not be the basis of Government liability even if negligent or an abuse of discretion, Congress has not immunized such activity if it exceeds the authority conferred. Liability, in such cases, is predicated not on a negligent or unwise policy determination, but on the failure of Government employees to conform to and act consistently with the authority delegated. We do not hold that the Government may be liable for policy determinations made by its officials. Rather, we hold only that the Government may be liable where its employees, in carrying out their duties, fail to conform to pre-existing statutory and regulatory requirements.

II. Negligence

Having concluded that this action is not barred by the discretionary function exception to the Tort Claims Act, we turn to the district court’s findings on the merits. The district court held that plaintiffs satisfied their burden of proving that the Government was negligent in releasing Lot 56 and that the Government’s negligence proximately caused Mrs. Griffin’s condition. With respect to the issue of causation the court found first, that Mrs. Griffin had polio; second, that Mrs. Griffin’s polio was of Type III variety; third, that the polio was caused by the vaccine; and fourth, that the vaccine ingested came from Lot 56.

We are unable to state that the findings of fact of the district court as to negligence²³ are erroneous.²⁴ The Government, however, contends that as a matter of law neither the statute, 42 U.S.C. § 262(d), nor the regulations promulgated thereunder, create a duty owed to individual members of the public. Citing the Restatement (Second) of Torts § 288(b),²⁵ the Government argues that the statutory language concerning the “safety, purity, and potency” of biological products contemplates at most a duty “vis a vis manufacturers” and the general public, but not running to individual consumers. We are aware of no legislative history, however, to support this narrow construction of the duty imposed by the statute.²⁶ Even assuming that Congress intended to provide for the public welfare, only if its intention was exclusively for the public welfare does § 288(b) of the Restatement apply. We are unable to conclude that Congress did not intend, at least in part, that the protections of § 262(d) and regulations extend to individuals as well as the public at large.²⁷

On the question of proximate cause, the district court found

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.

On appeal the Government does not contest Finding 20. However, the Govern7 ment forcefully contends that the court erred in finding that Mrs. Griffin ingested a dose of vaccine from Lot 56. We have independently examined the documentary evidence introduced on this issue. We are satisfied that plaintiffs have met their burden of proving by a preponderance of the evidence that Mrs. Griffin ingested a dose of vaccine from Lot 56. Plaintiffs’ evidence adequately established the presence of Lot 56 in Montgomery County at the time Mrs. Griffin ingested her dose of Sabin Type III vaccine. Although the evidence also shows the presence of vaccine produced from an additional lot, Lot 71, in perhaps greater quantity than Lot 56, we conclude, for the reasons stated by the district court, 351 F.Supp. at 18-23, that the trier of fact could properly find that Mrs. Griffin ingested vaccine from Lot 56 rather than Lot 71.

For the foregoing reasons the judgment of the district court on the question of liability will be affirmed.

III. Computation of Damages

We have reviewed the district court’s assessment of damages and, contrary to the Government’s position, “we cannot say that the award was in any way shocking, unfair or biased.” The hospital and medical bills as of the time of trial in July 1972 totaled $89,223.25. After considering plaintiffs’ expert’s calculations the court awarded $421,581.-00 as future medical expenses, based on a 25-year life expectancy and a very modest inflation factor of 2y2 percent,^27A and $49,142.40 as Mrs. Griffin’s impairment of future earning capacity. Thus, special damages alone totaled $559,946.65.

After finding that Mrs. Griffin’s sensitivity to pain has not been impaired by her condition; that she suffers “excruciating” pain from time to time; that she is mentally alert to her condition; that as a quadriplegic she is completely dependent on other persons, even for her bowel and bladder functions; and realizing that she had a 25-year life expectancy, the district court awarded $1,200,000 for her pain and suffering. In addition the court awarded Mr. Griffin $300,000 for past and future loss of consortium.

As Judge Hastie stated in Frankel v. Heym, 466 F.2d 1226, 1228 (3d Cir. 1972):

The problem here is inherent in any effort to translate . . . catastrophic human loss . . . into money damages. In this process systematic logic is not helpful and precision is not achievable.

The district court’s dispassionate opinion reflects the relevant considerations that led to its determination of damages. We cannot say that the award was in any way shocking, unfair or biased.

IV. Joint Tortfeasor Release

The Government maintains that the district court erred in refusing to reduce the plaintiffs’ recovery by 50 percent in accordance with the joint tortfeasor release executed by the Griffins in favor of Pfizer.²⁸ Noting that the release expressly provides that

[s]hould it appear that two or more persons or entities are jointly or severally liable in tort for the alleged injuries to wife-plaintiff, the considerations for this Release shall be received in reduction of the total damages recoverable against all the other tort-feasors to the extent of the pro-rata share of the said Pfizer Inc.,

the Government contends that under the Uniform Contribution Among Tort-feasors Act,²⁹ enacted in Pennsylvania July 19, 1951, 12 P.S. §§ 2082-2089, any judgment obtained by the Griffins against the United States must be reduced by the pro rata share of Pfizer, or 50 percent.³⁰ The district court, relying on Pennsylvania case law,³¹ held that because the Government never joined Pfizer as a third party defendant to determine Pfizer’s joint tortfeasor status, the Griffin-Pfizer agreement was legally ineffective to reduce the judgment.

As the district court points out, Pennsylvania has established the rule that in order to give effect to a “pro rata share” clause in a release, the joint tort-feasor status of the settling party must be judicially determined.³² Pennsylvania has apparently adopted this position to avoid allowing a defendant who may be the sole tortfeasor to obtain the benefit of the release. Eckels v. Klieger, 205 Pa.Super. 526, 210 A.2d 899 (1965). Thus, absent a determination of joint tortfeasor status, the benefit of the settlement inures to the plaintiff. This court has applied the Pennsylvania rule in Mazer v. Lipshutz, 360 F.2d 275 (3d Cir. 1966), where in a medical malpractice action against a surgeon we reverse the district court’s reduction of plaintiff’s verdict by the amount paid to the plaintiff by a hospital for its release.

The release executed in the instant case, however, differs from the releases in Davis, supra note 31, and Mazer in one crucial respect. The Griffin-Pfizer agreement provides:

In order to avoid inconvenience and expense to the released party, Pfizer Inc., in any action in which the said Pfizer Inc. is or may be a defendant or third party defendant together with other alleged tortfeasors, it is further agreed by us that any verdict rendered against the other alleged tortfeasors shall be reduced by the pro-rata share of the party released herein, Pfizer Inc., and any judgment entered on said verdict shall be in the amount of the verdict reduced by the pro-rata share of the party released herein, whether or not the released party herein was in fact a joint tort-feasor. This provision is intended to obviate the necessity and expense of having the released party herein remain a party on the record and obliged to participate at its expense in a trial merely for the purpose of determining if in fact it was a tort-feasor so as to entitle the other tort-feasors to a pro-rata reduction of any verdict. However, this provision in no way constitutes an admission of liability by the party released herein, Pfizer, Inc. [Emphasis supplied.]

This language constitutes an express waiver by plaintiffs of the benefits of the Davis holding. The Pennsylvania Supreme Court has stated:

To constitute a waiver of a legal right, there must be a clear, unequivocal and decisive act of the party with knowledge of such right and an evident purpose to surrender it. Kahn v. Banc America-Blair Corp., 327 Pa. 209, 193 A. 905 (1937); Cole v. Philadelphia Co., 345 Pa. 315, 26 A.2d 920 (1942).

Brown v. City of Pittsburgh, 409 Pa. 357, 186 A.2d 399, 401 (1962). That court has also stated:

While ordinarily the question of waiver is a question of fact for a jury . yet, where the only evidence as to waiver is a writing, its construction and interpretation and whether or not it constitutes a waiver is a question of law for the court.

Hanover Construction Co. v. Fehr, 392 Pa. 199, 139 A.2d 656 (1958). The above-cited clause of the. Pfizer-Griffin agreement is unambiguous. The Griffins have conceded the joint tortfeasor status of Pfizer for the purpose of determining damages in the case of a judgment tortfeasor. It is obvious that the release was drafted with Davis specifically in mind. Having waived their right to insist upon a judicial determination of Pfizer’s joint tortfeasor status, the Griffins may not object to a reduction of their judgment against the United States to the extent of Pfizer’s pro rata share.

The district court refused to give effect to the foregoing language in the release because it believed it inoperative until Pfizer was actually made a party to the Griffin v. United States action. The language of the provision on this point is ambiguous.³³ It is unclear whether this language contemplates that Pfizer be in fact made a defendant of record. Even if we construe the language as contemplating a requirement of Pfizer’s joinder, however, had the United States attempted to join Pfizer, the latter, under the terms of the release, could have obtained a dismissal. We see no reason to deny to the United States the benefits of the release because of its failure to make- a pro forma motion for joinder.³⁴

The case will be remanded for proceedings consistent with this opinion.

1 . 28 U.S.C. § 1346(b) provides:

Subject to the provisions of chapter 171 of this title, the district courts . . . shall have exclusive jurisdiction of civil actions on claims against the United States, for money damages, accruing on and after January 1, 1945, for injury or loss of property, or personal injury or death caused by the negligent or wrongful act or omission of any employee of the Government while acting within the scope of his office or employment, under circumstances where the United States, if a private person, would be liable to the claimant in accordance witli the law of the place where the act or omission occurred.

2 . See note 8 infra for text of statute.

3 . The district court opinion on liability and computation of damages is reported at 351 F.Supp. 10 (E.D.Pa.1972). The court’s opinion on the effect of the joint tortfeasor release is reported at 353 F.Supp. 324 (E.D.Pa.1973).

4 . According to newspaper accounts, Sabin Type II vaccine was to be administered on December 8, 1963. Mrs. Griffin did not receive a dose of Sabin Type II.

5 . As a result of tbe release, tbe Griffins voluntarily dismissed the two state court actions against tbe medical society.

6 . Tbe hierarchy of tbe Department of Ilealtb, Education and Welfare was stated by tbe district court as follows :

One of tbe divisions of tbe Department of Health, Education and Welfare was tbe Public Health Service, which was beaded by tbe Surgeon General. Tbe Public Health Service was divided into four divisions, one of which was called tbe National Institutes of Health, beaded by a director who figures but slightly in the case at bar. One of tbe National Institutes of Health was called tbe Division of Biologic Standards (D.B.S.).

351 F.Supp. at 25.

7 . 42 C.F.R. § 73.114(b) (1) (iii) provides:

(iii) Determination of neurovirulence. At tbe conclusion of tbe observation period comparative histopathological examinations shall be made of tbe lumbar cord, cervical cord, lower medulla, upper medulla and meseneapbalon of each monkey in tbe groups injected with virus under test and those injected with the NIH Reference Attenuated Poliovirus, except that for animals dying during tbe test period, these examinations shall be made immediately after death. Tbe animals shall be examined to ascertain whether tbe distribution and histological nature of tbe lesions are characteristic of poliovirus infection. A comparative evaluation shall be made of the evidence of neurovirulence of tbe virus under test and tbe NIII Reference Attenuated Poliovirus with respect to (a) tbe number of animals showing lesions characteristic of poliovirus infection, (b) tbe number of animals showing lesions other than those characteristic of poliovi-rus infection, (e) tbe severity of tbe lesions, (d) tbe degree of dissemination of tbe lesions, and (e) the rate of occurrence of paralysis not attributable to tbe 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 neuro-virulence of the test virus pool does not exceed that of the NIH Reference Attenuated Poliovirus. [Emphasis supplied.]

The “NIH Reference Attenuated Poliovi-rus” was designated by regulation to be Type I poliovirus. 42 O.E.R. § 73.111. It was referred to by the district court as NA-2 and will be referred to herein as the “reference strain.”

The regulation specifically refers to “virus pool” under test. Both the district" court and the briefs filed with this court treat “virus pool” and “lots” interchangeably. We continue this treatment.

42 O.F.R. § 73.114(b)(1) (iii) is a regulation promulgated by the Surgeon General pursuant to the Public Health Service Act, 42 U.S.O. § 201 et seq.; see 42 U.S.O. § 262(d).

8 . 28 U.S.O. § 2680(a) provides :

The provisions of this chapter and section 1346(b) of this title shall not apply to — (a) Any claim based upon an act or omission of an employee of the Government, exercising due care, in the execution of a statute or regulation, whether or not such statute or regulation be valid, or based upon the exercise or performance or the failure to exercise or perform a discretionary function or duty on the part of a federal agency or an employee of the Government, whether or not the discretion involved be abused.

9 . See the opinion of the district court, 351 F.Supp. at 23-26 for a discussion of the history of the development of the killed-virus and live-virus polio vaccines.

10 . At trial “neurovirulence” was defined as follows:

Neurovirulence, by definition, would be virulence or effect on the central nervou« system. As relates to poliomyelitis, it would indicate histologic changes in the central nervous system, which could be related to the pathologic entity known as polioencephalomyelitis.

In the context that one would talk about it here, this would be specifically for a monkey.

11 . As the Government points out in its brief, the regulations promulgated pursuant to 42 U.S.C. § 262(d), 42 C.F.R. §§ 73.110-73.118, prescribe various tests which the manufacturer must complete prior to the submission of a lot to DBS for release. The regulations further provide the manufacturer must submit “all protocols relating to the history of manufacture of each lot of vaccine, and the results of all tests performed.” 42 C.F.R. § 73.116(g)(1). Although the Director of DBS, Dr. Murray, testified at trial that “the tests by the manufacturer were considered as the test on which the product should be released,” the regulations provide for the submission of samples of the vaccine to DBS in addition to protocols and test results. The testimony was uneontradicted and the district court'found, 351 F.Supp. at 27, that in the case of Type III test virus, because of the lack of reliability of the manufacturers’ test results, DBS performed its own monkey neurovirulenee tests on every sample of vaccine submitted.

12 . DBS performed all neurovirulenee tests of the reference strain. A sample of the reference strain was inoculated periodically into groups of monkeys, typically in 30-monkey batches. Thus, for example, at the time DBS tested Lot 56 for neurovirulenee, it had a cumulative experience with the reference strain of 305 monkeys.

At the time the neurovirulenee test of any given test lot was to be compared with the neurovirulenee of the reference strain, the neurovirulenee of the latter was determined by the cumulative experience of the reference strain as of that date.

13 . After a prescribed period following inoculation, monkeys were sacrificed and microscopic examinations were made of sections of the monkey’s spinal cord and brain. Lesions that appeared in the slides were graded for “severity” and “spread.” “Spread” is the grade of lesion that occurred when the virus spread down to the site furthest removed from the inoculation site. “Severity” is the grade of lesion that occurred in the area closest to the inoculation site. Lesions were graded from grade 1 (minimal) to grade 4 (severe), in accordance with the damage observed to the anterior horn cells. See 351 F.Supp. at 16 (finding 33).

14 . The Director of DBS testified that

[W]e considered the number of monkeys that were involved with lesions of any kind as being the most significant point of comparison . . . . This was a position which had been taken right from the onset of the evaluation of neurovirulenee findings.

15 . Plaintiffs’ strict interpretation appears to have been rejected by the district court as well. See 351 F.Supp. at 27. The court noted that DBS was required to give enumerated factors “reasonable” weight.

16 . Assuming, arguendo, that the regulation authorized DBS to weight the ehronologieal results of the reference strain tests, such weighting would not require the conclusion that approval of lots under regulation, 73.115 (b) (1) (iii) was a “discretionary function.” Such weighting would reflect a decision that, for example, later tests, due1 to refinement of test technique, were a more accurate measure of reference strain neurovirulenee. Again, such a decision involves not public policy but a scientific judgment as to the accuracy of test results.

16A . The function of DBS was not, as stated in the dissent, to determine whether lots were “safe for release to the public.” Its specific duty under the regulation was to ascertain whether the neurovirulenee standard of the regulation had been met. If so, the lot could be released. This function of comparing the test lot with the reference strain did not “constitute the planning of the method and procedures by which DBS would undertake to perform regulatory duty.” Dis-seating opinion at 7. The method and procedures to be followed by DBS were spelled out specifically and in detail by the regulations promulgated by the Surgeon General with the advice of his committee.

17 . In Hendry the court held that the “discretionary function” exception did not bar judicial review of a claim based upon the allegedly negligent suspension of a license. The licensing determination depended upon psychological tests administered by Government psychiatrists.

Our position is likewise supported by those decisions holding the exception inapplicable to claims based upon allegedly negligent treatment rendered by Government physicians. See, e. g., Costley v. United States, 181 F.2d 723 (5th Cir. 1951) (injection of harmful substance causing paralysis). See also United States v. Gray, 199 F.2d 239 (10th Cir. 1952) (determination to leave deranged patient unattended); Fair v. United States, 234 F.2d 288 (5th Cir. 1956) (release of psychiatric patient); White v. United States, 317 F.2d 13 (4th Cir. 1963) (allowance of freedom of movement to psychiatric patient).

See Note, The Federal Seal of Approval: Government Liability for Negligent Inspection, 62 Geo.L.J. 937, 960 (1974).

18 . A comparison of Lot 56 test results with DBS’ experience with the reference strain is indicated by the following chart :

Lot 56 Reference ^*

Number of Monkeys with Lesions 4 ' 2.4

Severity ^**

Grade 4 2 0

“ 3 1 .1

“ 1, 2 1 2

Spread **

Grade 4 0 0

“ 3 3 .1

“ 1, 2 1 2

Paralyzed monkey 1 0

* The reference strain in fact consisted of 305 monkeys. In the chart all reference strain test results have been divided by 10 to allow comparison with the 30-monkey test lot results.

** For the significance of grade scores see note 13 supra.

On their face, the results of the Lot 56 test showed that Lot 56 neurovirulence exceeded that of the reference strain. The Government argues that DBS regarded the criterion of number of monkeys with lesions as the most significant evidence of neurovir-ulence, and that under this test, the overall Lot 56 test results were not adverse. However, our examination of the chart shows that the results of the test lot exceeded those of the reference strain with respect to all criteria.

See also note 23, infra.

19 . A DBS official testified on direct examination as follows:

Q. At the time that the DBS made these tests on Lot 56 of the Pfizer Type 3 vaccine, did you evaluate those results?

A. Yes.

Q. And what conclusion or recommendation did you make?

A. These test results were evaluated and considerable discussion ensued between Dr. Murray, Dr. Hottle, and myself, because there was in the intrathalamic test one paralyzed monkey, and it was agreed that probably this was within biologic variation, and that since the total number of animals with polio was four on thirty, which was not that different from the total that might be expected from the reference, and in addition, this was one of five of a consistency series, the four previous lots having been satisfactory and released, that this vaccine lot was satisfactory for release.

The official’s testimony on cross-examination reveals that DBS discounted the unfavorable results obtained in the Lot 56 tests because of the factor of “biological variation” :

Q. Do you consider a score of two 4’s and a paralysis and a 3 and a 2 being not in excess of that NA-2 score?

A. Mr. Adler, as I have said before, it gave me pause, it gave us all pause. We considered it and finally made the judgment that it was within biological variation, particularly based on the fact that in the totality of lesions as demonstrated in the chart it equalled many of the replicate tests of NA-2.

20 . The regulation expressly provided that “paralysis not attributable to the mechanical injury resulting from inoculation trauma” is to be considered evidence of neurovirulence. 42 C.F.R. § 73.114(b)(1) (iii) (e). Only paralysis that,DBS could establish was due to inoculation trauma could be disregarded. We are unaware of any contention or jiroof that the symptoms of the paralyzed monkey in Lot 56 were due to inoculation trauma.

Further, § 73.115(a) provides:

Repeat tests. Tests may be repeated when it is demonstrated that the results were due to faulty test techniques.

Dr. Murray testified that DBS considered that:

[T]he thrust of this statement is that all tests are to be considered as being meaningful, and cannot be disregarded unless it can be demonstrated that the failure of the test was due to a faulty test technique. [Emphasis supplied.]

21 . Dr. I. S. Danielson of Lederle Laboratories, a manufacturer of vaccine, specifically referred to “biological variation” in suggesting that the Surgeon General relax the restrictive language of the proposed regulations. He suggested:

As in all biological systems, some variations are expected to occur from test to test on the same preparation, let alone different preparations. We, therefore, suggest that the wording [of § 73.110 (a regulation dealing with neurovirulence of “seed virus” in language essentially parallel to that of § 73.114(b) (1) (iii)) ] be as follows: “neurovirulenee in Macaca monkeys is not significantly different from that of the NIH Reference Attenuated Poliovirus.” [Emjdiasis supplied.]

He also suggested that the concluding language of § 73.114(b) (1) (iii) be revised to read:

If an analysis of the test results demonstrates that the neurovirulence of the test vaccina does not exceed the range established for the NIH Reference Attenuated Poliovirus. [Emphasis in original.]

Dr. Stones, chief virologist of Pfizer, Ltd., proposed that the concluding language of § 73.114(b) (1) (iii) be revised to read: “does not significantly exceed that of the NIH Reference Attenuated Poliovirus.” [Emphasis in original.] Dr. Sabin suggested that the regulations provide that “the virus under test is satisfactory if the number of monkeys exhibiting persistent paralysis ... is in the same range as that obtained by the same operator using the same technique of inoculation with the NIH Reference Attenuated Po-liovirus.” [Emphasis in original.]

22 . The cases cited by the Government on discretionary function, e. g., Weinstein v. United States, 244 F.2d 68 (3d Cir. 1957); Schmidt v. United States, 198 F.2d 32 (7th Cir. 1952), do not refute this position. Although some language in our decision in Mahler v. United States, 306 F.2d 713, 723 (3d Cir. 1962), can be construed as inconsistent with our holding, we believe iliahler is distinguishable. Plaintiff’s contention in Mahler, unlike the instant case, sounded in negligence. See 306 F.2d at 724. The court in Mahler did not inquire into or find that that statute specifically required the official to do that which he allegedly failed to do.

We add that this action is not barred by the first clause of § 2680(a) first, because plaintiffs are not challenging the validity of a statute or regulation, see Dalehite, 346 U.S. at 32, 73 S.Ct. 956, and second, because our conclusion concerning the disregard of mandatory regulatory commands precludes a finding of “due care.”

23 . The district court found:

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.

351 F.Supp. at 27.

24 . See note 18 supra.

25 . § 288(b) provides:

When Standard of Conduct Defined by Legislation or Regulation Will Not Be Adopted The court will not adopt as the standard of conduct of a reasonable man the requirements of a legislative enactment or an administrative regulation whose purpose is found to be exclusively

* * ❖

(b) to secure to individuals the enjoyment of rights or privileges to which they are entitled only as members of the public.

26 . In Mahler v. United States, 306 F.2d 713 (3d Cir. 1962), cited by the Government, we held only that where the legislative history of a statute providing for inspection of federally funded roads conclusively showed that such inspection was solely for the purpose of protecting the government treasury, the statute did not create a duty to travellers of the roads. Whether a statute that provided for government approval of highway construction plans that were “conducive to safety” created a duty to travellers was expressly not decided. 306 F.2d at 723.

27 . Cf. United States v. Dotterweich, 320 U.S. 277, 282, 285, 64 S.Ct. 134, 88 L.Ed. 48 (1943); United States v. Sullivan, 332 U.S. 689, 696, 68 S.Ct. 331, 92 L.Ed. 297 (1948); 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).

As the district court wrote, “it is clear that the conduct of DBS 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).” 351 F.Supp. at 34. See also Ridley v. Boyer, 426 Pa. 28, 231 A.2d 307 (1967).

27A . The allowance by the district court for the inflationary factor, an insignificant part of the total damage award, is not challenged by the Government. Therefore, we express no views as to the propriety of its inclusion in the damages awarded. See Hoffman v. Sterling Drug, Inc., 485 F.2d 132, 143-144 (3d Cir. 1973).

28 . The release is set forth in full in the opinion of the district court. 353 F.Supp. at 325-326.

29 . 12 P.S. § 2085 provides:

A release by the injured person of one joint tortfeasor, whether before or after judgment, does not discharge the other tortfeasors unless the release so provides, but reduces the claim against the other tortfeasors in the amount of the consideration paid for the release or in any amount or proportion by which the release provides that the total claim shall be reduced if greater than the consideration paid.

12 P.S. § 2086 provides :

A release by the injured person of one joint tortfeasor does not relieve him from liability to make contribution to another tortfeasor, unless the release is given before the right of the other tortfeasor to secure a money judgment for contribution has accrued and provides for a reduction to the extent of the pro rata share of the released tortfeasor of the injured person’s damages recoverable against all the other tortfeasors.

30 . In the district court the Government maintained that the Griffin-Pfizer agreement in effect released two joint tortfeasors, Pfizer and the Montgomery County Medical Society, and that therefore the judgment should be reduced by two-thirds. This position was rejected by the district court on the ground that the original release affected Pfizer only. The Government does not press this contention on appeal and we therefore do not rule on it.

31 . Davis v. Miller, 385 Pa. 348, 123 A.2d 422 (1956).

32 . Cf. Swigert v. Welk, 213 Md. 613, 133 A.2d 428 (Md.1957). But cf. Layne v. United States, 460 F.2d 409, 411 (9th Cir. 1972).

33 . The provision states: “in any action in which the said Pfizer Inc. is or may he a defendant . . . . ” [Emphasis supplied.]

34 . Moreover, the transcript reveals that on the opening date of trial, counsel for the Government attempted to raise the issue of the effect of the release by entering into a stipulation regarding the existence and amount of the Griffin-Pfizer settlement. The Government’s position with respect to that settlement was stated in clear terms:

It would be our position, Your Honor, that if there were a judgment in this case, then, depending on the state law, the judgment would either be reduced by the amount of the settlement or — and I believe this to be the Pennsylvania law — it would be reduced by half.

Plaintiffs’ counsel objected to introducing the matter at this stage of the proceedings.