Long v. American Red Cross , 145 F.R.D. 658 ( 1993 )

OPINION AND ORDER

KEMP, United States Magistrate Judge.

This is a personal injury action involving the collection and transfusion of HIV-positive blood. According to the complaint, blood containing the AIDS virus was donated to the American Red Cross and, in September, 1984, transfused into plaintiff Judy L. Long. She died on December 21, 1992. Her husband, Gary Long, is also a plaintiff in this action. The -matter is currently before the Court to determine whether the Red Cross can be compelled to identify the donor of the blood at issue so that the plaintiffs can find out more about him.

Recently, the Red Cross advised the Court that the donor is now deceased. The basic issue to be decided is whether the plaintiffs’ need for information which is inaccessible to them unless they learn the identity of the donor outweighs the Red Cross’ interest in preserving the adequacy and safety of the nation’s blood supply, and the privacy interest of the donor or his family in keeping that information completely confidential. For the reasons that follow, the plaintiffs’ motion to compel discovery will be conditionally granted.

I.

Before beginning the discussion and analysis of this concededly difficult discovery issue, it may be helpful to explain why the particular analytical framework contained in this opinion has been selected. Despite the fact that the parties advance public policy arguments dealing with the discovery needs of injured parties, the privacy rights of blood donors, and the Red Cross’ desire to maintain an adequate and safe supply of volunteer-donated blood, what the parties have requested, and are entitled to, is a judicial decision.

It cannot be questioned that judicial decisions may be influenced legitimately by matters of public policy. It is also the case, however, that judicial decision-making differs from legislative or executive decision-making in at least one important respect: judges decide individual cases, or rule on motions presented within the framework of those cases. Each case, and each motion, has a unique set of facts which forms the basis for the judicial decision. Further, the level of proof required to establish a proposition as factual for purposes of a judicial decision is very different from the level which is required for purposes of making a legislative judgment. If judges stray too far from these basic propositions, and base their decisions on sweeping generalizations about public policy and on thinly-supported factual assertions, judges tend to become legislators. However, judicial decisions of this type are neither made in response to the will of the electors, nor subject to the type of debate and compromise which typically accompanies legislative pronouncements. It is important, therefore, to emphasize that the decision reached in this opinion is not an attempt to settle, or even to address, the larger question of whether, as a matter of policy, the right of a recipient of tainted blood to learn as much as possible about what happened when that blood was collected should always (or never) be given priority over the interests of either the collecting organization or the donor in keeping a part of that process confidential. Rather, “[i]n these tragic cases there is no hard and fast rule on when access to a donor should be allowed. These cases are extremely fact-sensitive, and the courts must decide each case by balancing the interests of the plaintiff, the donor, and society.” Snyder v. Mekhjian, 125 N.J. 328, 593 A.2d 318, 334 (1991) (Garibaldi, J., dissenting). The within decision, therefore, represents the Court’s best attempt to arrive at an appropriate balancing of these interests on the specific facts of this case.

Because of the need to focus upon the specific facts of this case, and because the facts which relate to each of the three distinct sets of interests involved are conceptually distinct, the opinion will be divided into a discussion of the facts and law that relate to (1) the plaintiffs' interest; (2) the privacy interest being asserted on behalf of the donor and the donor’s survivors; and (3) the interests of the Red Cross in providing a safe and adequate blood supply through the use of volunteer donors. While it might appear, at first blush, that some of these interests are not particularly fact-specific, a review of other decisions reveals that, even as to issues involving the safety and adequacy of the nation’s blood supply, the factual record in this case may differ significantly from the factual record (or lack of a factual record) in other cases. Consequently, the Court will now conduct a separate analysis of each of these three interests.

A. The Plaintiffs’ Interests.

On a general level, the interest of an injured plaintiff in discovering information which is relevant, as that term is defined in Fed.R.Civ.P. 26, to any cognizable legal theory on which recovery for his or her injuries may be had is an interest which has long been recognized by the judicial system and which may well be implicit in the Constitution. Certainly, under the Federal Rules of Civil Procedure, there is a presumption that any party to a lawsuit, in furtherance of either a claim or a defense, may obtain discovery of any information which has some identifiable relationship with the merits of that claim or defense. At least one court has suggested that this is part of the general right of access to the courts, which is a right of constitutional dimension. See Doe v. Puget Sound Blood Center, 117 Wash.2d 772, 819 P.2d 370 (1991).

Even cases which have denied the type of discovery being sought here have described the plaintiff’s interest in pursuing discovery as an “important” interest. See Doe v. American Red Cross Blood Services, 125 F.R.D. 646, 654 (D.S.C.1989). These general propositions, however, do little to advance the analysis in any individual case beyond identifying the long-standing recognition of the right to request and to pursue relevant information during the course of discovery. Since defining the dimensions of that entitlement in a particular case requires balancing the interests of those seeking discovery and those opposing it, it is important to determine in this case exactly how important the requested information is. That depends, in turn, on the theories which plaintiffs have advanced in this case, and the type of information to which these plaintiffs have already been given access.

First, the Red Cross has apparently conceded that Judy Long contracted AIDS from a blood transfusion, and that the transfused blood was supplied by the Red Cross. Thus, the plaintiffs do not have to show who the donor was in order to show that his blood, collected and supplied by the Red Cross, was the source of her infection. However, plaintiffs contend that their need for the information is not simply to demonstrate that the tainted blood came from the Red Cross, but to demonstrate that, but for the negligence of the Red Cross, that blood would never have been supplied to the hospital where Judy Long received her transfusion.

Plaintiffs’ negligence claim is based on two distinct theories. First, plaintiffs assert that, had the Red Cross employed what is known as “surrogate testing,” it might have discovered that the donor’s blood contained the AIDS virus. As described by plaintiffs’ counsel at a conference held with the Court on January 7, 1993, there were several “surrogate” blood tests available at the time of this particular donation which did not test directly for the presence of HIV, but produced positive results in 80% of the cases where the blood came from an HIV-positive donor. The experts can debate whether the Red Cross was negligent in failing to use a surrogate test given what it knew about AIDS and the spread of the AIDS virus in 1984. Information about the donor is not crucial to that aspect of this theory. However, if the jury concludes that surrogate testing should have been used, but also concludes that it was only about 80% accurate, the jury could find for the Red Cross by reasoning that the blood in question might not have reacted positively to the surrogate test, and that the failure to administer a surrogate test was therefore not a proximate cause of Judy Long’s injury. Indeed, one could reasonably expect the Red Cross to advance this precise argument at trial.

The most effective rebuttal of that argument would be evidence that the donor had his blood tested in this manner at or about the time he donated to the Red Cross, and that it tested positive. It is difficult to see how plaintiffs cbuld gather evidence on that issue other than by being allowed to pursue discovery about the donor which goes beyond the information which the Red Cross has supplied. For example, there may be doctor’s or hospital records that contain this information, or a friend or relative might have a record or a recollection of it. And, since it would ordinarily not be in keeping with the adversary system to permit or to require the Red Cross to pursue this line of inquiry for plaintiffs' benefit, disclosure of identifying information concerning the donor would seem to be the only way plaintiffs can further pursue this aspect of their first theory of liability.

Plaintiffs’ second theory of negligence is that the screening procedures followed by the Red Cross were inadequate. The Red Cross has provided plaintiffs with information concerning the screening procedure it used with this donor, including the pamphlet it distributed which described people who were at high risk for AIDS and instructed donors not to give blood if they believed they were high-risk individuals. The Red Cross has also provided plaintiffs with a list of the questions which were asked to the donor in question, and documents described as “the health history card from the implicated donation and from all prior and subsequent donations of the donor,” and “laboratory screening records from the implicated donation and from all prior and subsequent donations.” The question is then, whether additional discovery about the donor—which, in this case, would necessarily mean discovery from persons who knew the donor or from records about the donor available from other sources—would assist the plaintiffs in pursuing their “negligent screening” theory.

Originally, plaintiffs had argued that they did not wish to be bound by the Red Cross’ representation that the screening process occurred in the way in which the Red Cross described it. For example, they argued that the donor, if he recalled the screening interview, would be able to shed light on whether questions in addition to those on the form were asked, or whether some questions on the form were not asked. The donor’s version of this process could then be used as the basis for a hypothetical question to an expert, with the possibility that the jury could conclude that the screening process differed from the way it is depicted on the Red Cross’ records and, on that basis, find both negligence and proximate cause.

Obviously, because of the donor’s death, direct discovery from the donor is no longer available, and plaintiffs, in all likelihood, have to base their theory of recovery on an expert’s opinion that a different type of screening process should have been used. Knowing information about the donor will not help plaintiffs prove that the screening process should have been more comprehensive. As with the surrogate testing issue, however, if the plaintiffs succeed in persuading a jury that the Red Cross was negligent with respect to the screening procedures it employed, and that more or different questions should have been asked, the jury could find for the Red Cross by reasoning that the donor’s response to those additional or differently worded questions would not have led the Red Cross to refuse the donation or discard the blood.

Clearly, the donor cannot testify about what his responses to these types of questions would have been. Plaintiffs suggest that friends or relatives of the donor might know facts which, at least by inference, would address this issue. For example, if the donor was known as an “open homosexual,” he might have been willing to answer questions about that aspect of his life which were not asked by the Red Cross.

The Red Cross has argued that any such information from friends or family members would be inadmissible hearsay. Some of that information might well be inadmissible hearsay. Some might not. See, e.g., F.R.Evid. 803(3) (statements concerning physical condition) and F.R.Evid. 803(19) (reputation concerning personal or family history). Ordinarily, however, the fact that information obtained during discovery is hearsay is not an objection to its discoverability. Rather, evidence is discoverable as long as it is reasonably calculated to lead to the discovery of admissible evidence. Mellon v. Cooper-Jarrett, Inc., 424 F.2d 499 (6th Cir.1970).

Determining what type of information is reasonably, as opposed to unreasonably, calculated to lead to the discovery of admissible evidence does involve a measure of theorizing as to what might occur if potentially inadmissible evidence is provided, but this is the type of judgment call that courts routinely make in dealing with discovery issues. Here, it cannot be said with any degree of assurance that no admissible or discoverable evidence would be obtained by plaintiffs on their negligent screening theory if they could gather information concerning the donor’s lifestyle, physical condition, or other such issues, from persons who knew the donor at the time the donation was made. Again, it would not seem that plaintiffs could pursue this aspect of their negligent screening theory without additional discovery.

Concluding that the plaintiffs would or might benefit from obtaining the discovery they have requested is only the first step in assessing the nature of their interest, however. Because there are competing interests to which the plaintiffs’ interest will be compared, the Court must also evaluate the likelihood that plaintiffs will benefit from more discovery, and the relationship of that benefit to the plaintiffs’ entire case. Thus, the Court must ask whether the type of information being requested is likely to lead to evidence which is central to the plaintiffs’ case, or merely peripheral, and how likely it is that they will learn anything of either central or peripheral importance if provided with the discovery requested.

The decisions cited to the Court by the parties—all of which have been read, and which, taken together, represent a very comprehensive discussion of the issues involved in this type of case—appear to be split on the importance of this type of information to the plaintiffs’ ability to prove negligence. For example, information about what the donor would likely have disclosed in response to different questions during the screening process has been described by one court as “highly pertinent” to the issue of negligence. Snyder v. Mekhjian, 244 N.J.Super. 281, 582 A.2d 307, 314 (1990), aff'd 125 N.J. 328, 593 A.2d 318 (1991) (per curiam). See also Doe v. Puget Sound Blood Center, supra, describing this type of discovery as being at “the very heart of plaintiff’s negligence claim.” Doe, 819 P.2d at 376. On the other hand, the plaintiffs’ interest in pursuing discovery of this type has also been described as “modest,” Doe v. University of Cincinnati, 42 Ohio App.3d 227, 232, 538 N.E.2d 419 (Franklin Cty.1988), or as evidence with “dubious” probative value. Rasmussen v. South Florida Blood Service, Inc., 500 So.2d 533, 538 (Fla.1987). Certainly, it is not as important here as in a case where there are multiple donors involved and the Red Cross does not concede that any of them were or could have been the source of the plaintiffs’ infection. Cf. Tarrant County Hospital District v. Hughes, 734 S.W.2d 675 (Tex.App.1987). Nevertheless, the Court is convinced that this information is of more than peripheral, slight, or modest value to the plaintiffs. For that matter, it is also of potential value to the Red Cross and to the public interest in the truth-finding process, depending upon what information is gathered. Whether the employment of different tests or screening procedures would have produced a different result is a vital link in the plaintiffs’ attempt to prove proximate cause. Knowing that the donor did, or did not, undergo a surrogate test in 1984 and knowing the results of any test conducted are preferable to speculation or even the drawing of reasonable inferences about such matters. Thus, the type of information being pursued by the plaintiffs is of significant value.

The answer to the other question—how likely it is that the discovery sought by plaintiffs will lead to admissible evidence— cannot, at the moment, be known. Thus, it is difficult to weigh this factor at all, other than to say that plaintiffs have met the “reasonably calculated to lead to admissible evidence” standard. Taking all of these matters into account, the Court concludes that these particular plaintiffs’ need for this information is an interest which cannot be brushed aside lightly by generalized concerns either about the safety or adequacy of the blood supply, or marginally protectable privacy interests of the deceased donor or his survivors. Thus, the analysis now turns to the strength and character of these two interests.

B. The Donor’s Privacy Interest.

In other cases where the issue of discovery from or about a blood donor has been raised, the Red Cross has been permitted to advance the donor’s- interest in not having his identity or other sensitive information revealed. This Court is persuaded that the Red Cross should be granted legal standing to do so. Most courts have also concluded that the donor has a legitimate privacy interest which is entitled to be given some weight in the analysis. The source of that privacy interest is variously described as stemming from either the United States Constitution, a state constitutional provision, from the common law, or from Fed.R.Civ.P. 26(c), which provides, in relevant part, that “[u]pon motion by a party or by the person from whom discovery is sought ... the court ... may make any order which justice requires to protect a party or person from annoyance, embarrassment, oppression, or undue burden or expense____” Most discovery issues which involve privacy or confidentiality concerns are dealt with under Rule 26(c) through the issuance of an order either “that the discovery not be had,” or that it be conducted only under guidelines set up by the court which are designed to safeguard the privacy or confidentiality interests being asserted.

No extended discussion of the source of the donor’s privacy interest is really necessary. What is important in this case is the extent to which the donor’s legally protectable privacy interests, from whatever source, would be infringed by permitting plaintiffs to undertake discovery, and, if unfettered discovery would infringe upon the donor’s privacy interest, whether that infringement can be adequately addressed by a protective order less drastic than “that the discovery not be had.”

The term “privacy” is used in the law to cover a number of conceptually distinct individual rights. For example, preventing others from portraying an individual in a “false light,” or stopping the unauthorized use of one’s likeness for commercial gain, have been characterized as privacy rights. See Spahn v. Julian Messner, Inc., 18 N.Y.2d 324, 274 N.Y.S.2d 877, 221 N.E.2d 543 (1966), vacated on other grounds, 387 U.S. 239, 87 S.Ct. 1706, 18 L.Ed.2d 744 (1967); Zacchini v. Scripps-Howard Broadcasting Co., 47 Ohio St.2d 224, 351 N.E.2d 454 (1976), rev’d on other grounds 433 U.S. 562, 97 S.Ct. 2849, 53 L.Ed.2d 965 (1977). Those aspects of the right to privacy are not involved in this case, however. Rather, there are three distinct privacy interests which might conceivably be invaded if the plaintiffs were allowed to proceed with the discovery they have requested. Those interests are: (1) the interest in not having information of a personal, sensitive, or embarrassing nature made known to third parties; (2) the interest in preventing any inquiry into extremely private matters, such as sexual relationships; and (3) the more generic “right to be let alone.” Olmstead v. United States, 277 U.S. 438, 478, 48 S.Ct. 564, 572, 72 L.Ed. 944 (1928) (Brandeis, J., dissenting). The Court will examine each of these interests in turn.

Society has generally recognized that the publication of truthful information concerning details of a person’s private life can, under certain circumstances, abridge a legally protectable right to keep that information private. See, e.g., Time, Inc. v. Hill, 385 U.S. 374, 87 S.Ct. 534, 17 L.Ed.2d 456 (1967). The basis for recognizing such an interest would appear to be, at least in part, a concern that the publication of such information may damage a person’s employment, family relations, or reputation and standing in the community. The potential for injury to a blood donor caused by publication of the fact that he suffered from AIDS appears to be the primary concern of those courts which have characterized this privacy interest as strong. See, e.g., Rasmussen v. South Florida Blood Service, Inc., 500 So.2d 533, 536, 537 (1987), describing the donor’s privacy interest as “the right to determine whether or not sensitive information about oneself will be disclosed to others,” and discussing “the possibility that a donor’s co-workers, friends, employers, and others may be queried as to the donor’s sexual preferences, drug use, or general life-style.” The Rasmussen court also noted that, in today’s society, being identified as a victim of AIDS is akin to having leprosy. See also Doe v. American Red Cross Blood Services, 125 F.R.D. 646 (D.S.C.1989); Doe v. University of Cincinnati, 42 Ohio App.3d 227, 538 N.E.2d 419 (Franklin Cty.1988).

Although individuals generally enjoy an expectation that private information of this type will not be disclosed to third parties without their consent, that expectation is typically lessened where an individual has engaged in activity which can lead to civil liability. For example, it would not seriously be argued that, in an automobile accident case, the defendant could not be asked about whether he or she was an alcoholic or was drinking on the day in question, even though being labeled an alcoholic carries with it a certain amount of social disapproval and may have other serious adverse consequences. Thus, a fairly compelling argument can be advanced that if someone knows or should know that his blood is unsuitable for donation and may cause injury to others if transfused into them, by donating blood the person gives up any legitimate expectation that information about his or her medical condition will not, at some point, be divulged at least to the person who received the blood, and possibly to others as well if it is necessary in order for the injured party to pursue a legal claim. Consequently, the existence of this type of privacy interest is not regarded as an absolute bar to discovery, but is rather an interest which is taken into account by fashioning an appropriate protective order limiting disclosure of the information, and providing that it will be used solely for purposes of the litigation. Conversely, it would be an unusual case where this kind of privacy interest is not susceptible of being accommodated by an appropriate protective order, although some courts have apparently concluded that information about AIDS is, at least for a living donor, of a sufficiently extraordinary nature to justify a total denial of discovery.

In this case, however, the donor is dead. At least one other court has squarely addressed that issue and has concluded that the donor’s interest in such matters as his or her reputation, job security, employability, availability of credit, and similar matters “would seem to disappear” when the donor dies. Doe v. Puget Sound Blood Center, 117 Wash.2d 772, 819 P.2d 370, 376 (1991). Certainly, the development of the law of privacy seems to reflect a societal judgment that such interests are entitled to either lesser, or no, recognition following the death of the individual.

Cases which deal with legal characteristics of the interest in avoiding publication of embarrassing or damaging information are helpful in defining the extent to which a deceased blood donor has such a privacy interest. Ohio, for example, does not permit a cause of action for invasion of privacy to survive the injured party’s death. “It is unquestioned ... that actions for invasion of privacy in Ohio are not descendible and lapse upon death." Reeves v. United Artists Corp., 765 F.2d 79, 80 (6th Cir. 1985), citing, inter alia, Young v. That Was the Week That Was, 423 F.2d 265 (6th Cir.1970). The District Court decision in the Young case, which was affirmed on appeal, notes that “[t]he right of privacy has always been treated by the courts as a personal right fashioned to protect an individual’s solitude,” and that the right “dies with” the individual. Young v. That Was the Week That Was, 312 F.Supp. 1337, 1340-41 (N.D.Ohio 1969), aff'd 423 F.2d 265 (6th Cir.1970). Many other jurisdictions recognize this limitation, see Cordell v. Detective Publications, Inc., 419 F.2d 989, 990 and n. 3 (6th Cir.1969), and this is the position taken by the Restatement of Torts 2d as well. See Loft v. Fuller, 408 So.2d 619, 622 (Fla.App. 1982), citing Restatement of Torts, 2d, § 652 I (1977). Thus, although this case does not directly involve the issue of whether the donor would have a legal claim for invasion of privacy that could be asserted against someone who published damaging information about him, the law’s reluctance to recognize such a claim is good evidence that the legal protection accorded to this type of privacy interest is lessened or eliminated once the donor has died.

The Red Cross cites Kiraly v. F.B.I., 728 F.2d 273 (6th Cir.1984), in support of its argument that the donor’s death is not of crucial significance. That case addressed the question of whether the Freedom of Information Act’s prohibitions against releasing information that would constitute an unwarranted invasion of privacy or disclose the identification of a confidential law enforcement informant were unenforceable after the informant’s death. See 5 U.S.C. § 552(b)(7)(C) and (D). That case, of course, involved interpretation of a statutory provision not at issue here, and held only that there was no automatic duty to disclose such information where the public interest was minimal, and that fear of retaliation against an informant’s family was an interest that survived the informant. The Freedom of Information Act, as interpreted by Kiraly, does not seem to be a Congressional pronouncement that deceased persons generally enjoy a broad range of privacy rights. Thus, this Court concludes that the donor’s privacy interest in preventing the publication of embarrassing or sensitive information is significantly diminished, if not extinguished altogether, by reason of the donor’s death.

The same would appear to be true with respect to any asserted interest in the privacy of matters of a sexual nature. First, it is not at all clear that, at least with respect to some types of sexual conduct, this right is distinct from the right to keep sensitive information about oneself out of the public domain. Compare Eisenstadt v. Baird, 405 U.S. 438, 453, 92 S.Ct. 1029, 1038, 31 L.Ed.2d 349 (1972) (right of privacy in matters related to conception of children) and Griswold v. Connecticut, 381 U.S. 479, 85 S.Ct. 1678, 14 L.Ed.2d 510 (1965) (right of marital privacy) with Bowers v. Hardwick, 478 U.S. 186, 191, 106 S.Ct. 2841, 2844, 92 L.Ed.2d 140 (1986) (state may constitutionally punish various types of consensual sexual activity, including homosexual activity). Even if this interest is distinct from the interest in preventing publication of embarrassing or sensitive information, and would extend to the mere inquiry about sexual habits, preferences, or relationships, it is difficult to see the rationale for protecting such an interest after a person is dead. Both the embarrassment of having such information revealed, and the concern that one lives in a society where the privacy of one’s sexual life is not respected, appear to be rights of a personal nature that lapse upon death.

The Red Cross argues, however, that the privacy interests of persons other than the donor are also at stake in this case. It asserts that there is the potential that if the donor’s identity is disclosed, close family members will be the subject of some inquiry concerning sensitive matters about the donor, and that such inquiries are an invasion of those family members’ own right to privacy. The short answer to that assertion is that the law does not generally recognize that any third party, even a close family member, has a legally protectable interest in keeping private information known about another family member, other than perhaps in the context of the marital privilege.

The only Ohio decision which the Red Cross cites, State of Ohio ex rel. The Vindicator Printing Co. v. Watkins, Case No. 91-T-4555 (Trumbull Cty.App., December 31, 1991), 1991 WL 280007, 1991 Ohio App. LEXIS 6414, provides little support for this argument. In that case, the court concluded that Ohio had not decided whether family members had such a right, and noted that the few other courts which seemed to recognize it had done so on the basis of state statutes which acknowledged such a right. Watkins, 1991 WL 280007, at *8-9, 1991 Ohio App. LEXIS 6414, at *23-25. The Red Cross’ other cited authority, Doe v. Borough of Barrington, 729 F.Supp. 376 (D.N.J.1990), to the extent that it would extend this privacy right to members of an AIDS victim’s “immediate family” (a phrase the court appeared to equate with “[tjhose sharing a household with an infected person,” see Doe, 729 F.Supp. at 385), is also of questionable relevance. At this point, it is unknown whether there are any members of the donor’s “immediate family” or persons with whom he shared a household who might be the subject of some type of inquiry. If there are, Rule 26 permits the Court to regulate discovery in such a way as to prevent embarrassment to others, and it would therefore appear that issues of this sort cannot meaningfully be addressed until it is ascertained who, if anyone, might be affected by a limited disclosure of the fact that the donor had AIDS, or an inquiry into that subject.

There is one other aspect of the privacy issue which should also be addressed. The Red Cross, and some of the decisions which it cites, such as Doe v. University of Cincinnati, supra, have focused on donors’ expectations of privacy which are created by the Red Cross’ representations to those donors that the information they disclose to the Red Cross will be kept confidential. The impact of those representations on blood donors generally, in terms of whether donors will be either more likely to lie or less likely to donate blood if those assurances are breached, is discussed below. The question addressed here is whether those assurances might also create a legitimate privacy interest independent of the type of information that is communicated pursuant to such an assurance of confidentiality.

For two separate reasons, this Court does not believe that the representations made by the Red Cross to the donor carry any significant weight when analyzing the nature or extent of the donor’s privacy interest. First, the donor is deceased, and if this is a separate privacy interest, it likely expired with the donor. Second, it is generally held that private agreements to keep information confidential are not enforceable against third parties if the information agreed to be kept confidential is not deserving of protection in the absence of a confidentiality agreement. Cf. Sharjah Investment Co. v. P.C. Telemart, 107 F.R.D. 81 (S.D.N.Y.1985). In other words, relevant, unprivileged, and non-confidential information cannot be withheld in the face of a legitimate discovery request simply because the party who possesses that information has agreed with the provider of the information not to disclose it. If disclosure of the information would breach some private agreement, the party in possession of that information may wish to seek a protective order in order to be able to defend a claim that the confidentiality agreement was breached. However, the fact that such an agreement exists would appear to carry little weight in terms of whether the Court should order disclosure of the information.

There is at least one exception to that general rule, however. If the information itself is not of a confidential nature, the Court may still give effect to the parties’ agreement not to disclose it if some other policy reason for doing so can be found. Thus, for example, parties who have agreed not to disclose the terms of a settlement may legitimately withhold that information from discovery, even if it might be relevant to some other litigated matter, based upon the Court’s view that the policy in favor of encouraging settlements is stronger than the inquiring party’s need to know the terms of a settlement. See Gaull v. Wyeth Laboratories, 687 F.Supp. 77 (S.D.N.Y.1988).

In this case, it may well be that the Red Cross made representations concerning confidentiality which directly induced the donor to provide the information which plaintiffs now seek, and that the public interest is served by allowing the Red Cross to make such representations. The existence of these assurances might support a claim by the donor against the Red Cross if the Red Cross thereafter voluntarily disclosed this information, but the fact that there may be some enforceable agreement between those parties should not be taken into account in determining whether the information is discoverable from the Red Cross. Rather, the Court must focus either on the confidential nature of the information itself, which the Court has done in its discussion of the donor’s privacy interests, or on any public policy which might convince the Court to enforce the agreement, a question addressed below. Consequently, since the Court concludes that the guarantee of confidentiality provided by the Red Cross cannot create a legitimate expectation that no disclosure of otherwise non-confidential information will be made unless public policy dictates that it be kept secret, the fact that the Red Cross made such representations to this donor need not be analyzed separately from those other issues.

C. The Red Cross’ Interests.

Perhaps the most difficult interest to assess is that advanced directly by the Red Cross. By acting as a collector of blood, the Red Cross serves a vital public function. Our society depends heavily upon the ready availability of safe blood for transfusion purposes. Thus, the Red Cross is rightfully concerned with any event, including the disclosure of information about its donors, which could hinder its ability to perform this vital public function.

Undoubtedly, because the maintenance of a safe and adequate blood supply might be viewed by a policymaker as an interest which, in the abstract, is so important that it outweighs the need for an individual plaintiff, or the need of plaintiffs collectively, to acquire information to pursue personal injury claims, some courts have considered this interest to be dispositive of the issue currently before this Court. Coleman v. American Red Cross, 130 F.R.D. 360 (E.D.Mich.1990), aff'd 979 F.2d 1135 (6th Cir.1992), is one example of a decision which concludes that the risk posed to the safety and adequacy of the nation’s blood supply is sufficient to overcome the plaintiff’s interest in discovery. Laburre v. East Jefferson General Hospital, 555 So.2d 1381 (La.1990) is another such decision. That court concluded that discovery of the type sought by the plaintiffs would be “a significant disincentive to voluntary blood donations,” and that “[i]f donors believed that there was any threat of their being made a party to a lawsuit or of an investigation into their personal lives because of their blood donation, they may not provide completely accurate information.” Laburre, 555 So.2d at 1384-85.

Other courts have not viewed the issue in quite the same way. For example, in Stenger v. Lehigh Valley Hospital Center, 386 Pa.Super. 574, 563 A.2d 531, 537 (1989), aff'd 530 Pa. 426, 609 A.2d 796 (1992), the court stated that “[w]e can perceive of no correlation between ... a limited ... protective order and a reduced number of blood donations____” Courts have also concluded that, to the extent that the threat of such disclosure discourages high-risk donors from making donations, allowing discovery will enhance the safety of the blood supply, and will also encourage collectors of blood like the Red Cross to act responsibly in screening donors. See Stenger v. Lehigh Valley Hospital Center, 530 Pa. 426, 609 A.2d 796 (1992); Snyder v. Mekhjian, 125 N.J. 328, 593 A.2d 318, 324 (1991) (per curiam) (Pollock, J., concurring).

One possible explanation of why courts have reached seemingly irreconcilable conclusions with respect to the question of how disclosure of this type of information will affect the safety and adequacy of the nation’s blood supply is that the issue is primarily a factual one, but the courts have not necessarily treated it as such. There are, of course, many areas in the law where the courts have created rules for decision based upon what appear to be logical conclusions about the way people will behave under certain circumstances. For example, certain privileges, like the attorney-client or physician-patient privileges, rest upon an assumption that people will be less likely to disclose fully their legal or medical problems to a professional if they know that such information can be freely disclosed to third parties. It is likely that these privileges were created without a significant amount of empirical proof that this assumption is accurate.

When reasonable legal minds can differ, however, as to how a group of people would react to a particular judicial decision, courts should rightly demand some quantum of proof of the consequences of the decision. Reasonable minds can and have differed on the extent to which the Red Cross’ concerns are well-grounded in fact. For example, it appears to this Court that it is just as likely that high-risk donors who become aware that the very information which makes them high-risk donors could be disclosed to third parties in the event that their blood causes someone else to get sick or to die will stop donating blood altogether, than that they will lie about their medical history in order to continue to donate blood. It would seem unusual that a person altruistic enough to be a volunteer blood donor would nevertheless selfishly conceal such vital information, thereby creating a significant risk that the recipient of his or her blood would become gravely ill or would die. It also seems possible that low-risk donors have few, if any, significant or sensitive details to disclose and would therefore be unfazed by the prospect that some of these details might become public knowledge in the event that some problem developed as a result of use of their blood. Obviously, in addition to questioning the validity of these assumptions, one can reasonably suggest that there is a “gray area” between these two extremes where some amount of lying, or some amount of unwillingness to donate blood, might occur. The size of that gray area, and its importance to the safety and adequacy of the blood supply, are, in this Court’s view, matters which cannot be derived by logical or legal analysis from the proposition that blood donor information may, under certain limited circumstances, no longer remain confidential. Rather, these are factual matters and must be determined with reference to the factual record in this case.

The Red Cross has submitted several affidavits or declarations on this issue. The first, Exhibit D to the Red Cross’ reply brief filed on December 11,1992, is a declaration of H. David Banks, Ph.D. Dr. Banks is a behavioral scientist employed by the Red Cross, and his Ph.D. is in human development. He was the principal investigator for a donor survey which was conducted in June and July, 1992.

According to Dr. Banks’ declaration, the purpose of his study was to determine what effect the release of identifying information would have on blood donors. Three hundred and sixty-one donors were surveyed and were asked to check one of five responses to certain questions. The questions asked the donors whether they were willing to (1) give accurate information about medical and personal history when giving blood; (2) take free additional blood tests; (3) provide more detailed personal information upon request; and (4) donate again during the next twelve months. The five available responses were (1) strongly agree; (2) agree; (3) not sure; (4) disagree; or (5) strongly disagree. According to Dr. Banks, the results of the survey showed that about 50% of the donors indicated that “they would be less willing to give accurate medical and personal history information if the Red Cross were required to release such information for use in litigation” and “another 25.6% were either unsure or indicated that they would not be willing to provide accurate information under those circumstances.” The donors surveyed were also less willing to agree to take additional tests if confidentiality were not assured, less willing to provide additional information under those circumstances, and less willing to donate blood again if the confidentiality policy were not in place. Dr. Banks’ conclusion was that there is a statistically significant difference between donor intentions with the current Red Cross confidentiality protection in place, and their intentions if this information could be released for use in litigation.

In addition to this declaration, the Red Cross also submitted an affidavit of A. William Shafer, M.D., the principal officer of the Red Cross' Southeastern Michigan Blood Services Region. The affidavit was originally filed in connection with a motion to dismiss the complaint in Coleman v. American Red Cross, Case No. 89-CV-71671-DT, a case in which, contrary to a court order, plaintiffs learned the identity of the donor and then announced an intent to sue him. An article concerning that matter appeared in the Detroit Free Press on August 27, 1992. According to Dr. Shafer, in the two weeks thereafter, the Southeastern Michigan region experienced a 13.2% drop in the number of units of blood collected when compared with the same two-week period during 1991. Dr. Shafer also noted that special appeals for blood during that time were less successful than special appeals the year before. His affidavit indicated that this effect was localized, and he expressed the “professional opinion” that the decline in blood donations was due, at least in part, to the Colemans’ discovery of the donor’s identity and their publicized intention to sue him.

In an effort to rebut these affidavits, plaintiffs submitted an affidavit from David Eisenman, the president of the Association for Improvement of Volunteer Blood Donation, who has been involved in the field of volunteer blood donation since at least 1972. His affidavit sets forth various positions he has held with governmental boards, such as task forces and research groups involved in blood banks and blood donations. Mr. Eisenman, after reviewing the statements of Dr. Banks and Dr. Shafer, expressed the “considered opinion,” that “granting the plaintiffs’ request for access to information about the blood donor implicated in the above-captioned matter is extremely unlikely to have any significant effect on the availability of blood for transfusion, and certainly not the major impact that Dr. Shafer and Dr. Banks would have the court expect.” He concludes that limited disclosure of identifying information under appropriate protective orders would cause only an insignificant percentage of volunteer blood donors to alter their donation patterns, that any such loss would be made up by additional donations, and that the United States donor pool is “so large and so resilient that even a large loss of donors for this reason would not cause life-threatening shortages of blood.” Mr. Eisenman also cites to a 1982 study which concluded that the actual number of blood donors relates directly to the limited need for blood rather than public resistance to blood donation, and he notes that the number of blood donors has actually increased during the time that the donor pool has been reduced by eliminating persons at high risk for AIDS.

In addition to these conclusions, Mr. Eisenman is critical of Dr. Banks’ survey instrument, asserting that it measured only opinions of donors rather than their actual intentions. He asserts that a colleague of his, who is a research professor in the Survey Research Laboratory and Professor of Sociology at the University of Illinois, is in agreement with his conclusions about the flawed nature of Dr. Banks’ survey. Finally, while he does not disagree with Dr. Shafer that the publicity about the Coleman case may have had a temporary detrimental effect on blood donations in southeastern Michigan, he asserts that any such shortage can be made up in other areas and that it is, at best, a short-lived problem. He also appears to agree with court opinions which have concluded that, to the extent that the field of donors is reduced, the reduction would come in high risk donors and would therefore contribute to blood safety.

In response to Mr. Eisenman’s affidavit, the Red Cross submitted a supplemental declaration of Dr. Banks. Dr. Banks states that his donor survey study will be published in a peer-reviewed professional journal, that standard research texts support his choice of responses for the survey, that experts on survey research from other universities provided similar support, and that Mr. Eisenman’s conclusions about the impact on the blood supply are “unsupported hypotheses____” Dr. Banks also adds his opinion that while his survey did not address whether donors’ intentions might change if such information could be released only after their death, “it is unlikely that donor responses would have significantly varied.” He believed that donors “do not want their families, friends, and neighbors subject to such questioning after their deaths.”

As one court has pointed out, in evaluating the type of affidavits which are submitted in support of the Red Cross’ argument, the Court must insure that these affidavits are competent, and are not simply unsupported opinions; “[t]he difficulty with these predictions is that they border on speculation about human conduct in reaction to a limited and restrictive discovery order.” Doe v. Puget Sound Blood Center, 117 Wash.2d 772, 819 P.2d 370, 379 (1991). Although the affidavits submitted by the Red Cross in this case cannot be branded as pure speculation, and may well reflect the results of an appropriately-designed public survey and the actual experience of the southeastern Michigan region following publicity about the Coleman case, it is this Court’s view, after closely scrutinizing the affidavits, that they are not terribly probative of the factual proposition that some disclosure of information concerning a volunteer blood donor will have a measurable or significant impact on the blood supply of the nation, or on the truthfulness of the persons donating blood.

First, the Court is not persuaded that Dr. Shafer’s affidavit proves anything beyond the fact that fewer blood donors came forward to donate blood in southeastern Michigan in the first two weeks of September, 1992 than they did in September, 1991. To draw an inference that this was related in some fashion to publicity concerning the Coleman case is to engage in the logical fallacy of post hoc ergo propter hoc. As Mr. Eisenman’s affidavit points out, and as plaintiffs in the Coleman case also argued, a much more scientifically-based study would be required in order to show that the drop in donations not only followed publicity concerning the Coleman case but was caused by it. The Coleman court, in its subsequent opinion dismissing that case as a sanction for violation of the court’s protective order, noted that “plaintiffs’ argument that further study is needed to ascertain the exact cause or causes for the drop in blood donations has merit.” Coleman v. American Red Cross, 145 F.R.D. 422, 430, (E.D.Mich.1993). Further, as the Eisenman affidavit points out, to the extent that a temporary and localized drop in the number of donors occurred, the possibility that increases in other areas or the “elasticity” of the donor pool will make up for a temporary shortfall still exists. Consequently, this evidence is not strongly probative of the Red Cross’ claim.

In contrast, the study performed by Dr. Banks can be characterized as scientifically well-grounded, at least on the basis of the record before the Court. The question is not so much the reliability of the study, however, as whether it provides an adequate basis for the conclusions which Dr. Banks has reached. A close examination of his declaration reveals that the link between his data and his conclusions is by no means iron-clad.

For example, Dr. Banks’ second set of survey questions, which were designed to find out whether donors would be more or less likely to donate blood or provide accurate information in the event that there might be some disclosure of this information to third parties, begins with the following sentence:

“If the American Red Cross was required to release your personal history information to lawyers and courts for use in lawsuits including public trials, would you agree or disagree with the following statements?”

That factual premise does not necessarily reflect the action which the Court is considering in this case. First, of course, it does not include any language such as “only after you have died.” Dr. Banks’ opinion to the effect that the results of the survey would not change if such language were included is, on this record, also an “unsupported hypothesis.” Second, it does not incorporate any reference to limited disclosures of either identity or information, disclosures under court supervision, or disclosures in the context of a lawsuit brought against the Red Cross for negligence in the screening process which has caused someone else to contract a potentially fatal disease. Rather, it might well have tapped into a general societal fear and dislike of lawyers and litigation, without being tempered by the specifics about the nature or purpose of the litigation. A blood donor might well have a different reaction to a question concerning disclosure if the purpose of the disclosure was to permit injured parties to bring suits against the Red Cross, and if one of the potential benefits of such a suit might be to cause the Red Cross to employ screening processes that would make the nation’s blood supply safer. Donors, it is presumed, do not act entirely out of unselfish motives, but also believe that some day they will be in need of transfusions. The omission of this type of information from the survey question suggests that it does not directly address the issue posed in this case, and that opinions which attempt to stretch results of the survey to fit this precise factual situation are not particularly well-supported.

The Court also notes that Dr. Banks has not provided the results of his survey other than in summary fashion. For example, his declaration states that approximately 50% of donors stated that they would be “less willing” to give accurate information if the Red Cross were required to release such information for use in litigation. In response to the question which incorporated the present confidentiality policy, 91% of donors “strongly agreed” that they would provide accurate information. It would appear, based upon Dr. Banks’ statement that “another 25.6% were either unsure or indicated that they would not be willing to provide accurate information” if the information could be released, that the 50% who are “less willing” are people whose response changed from “strongly agree” to “agree.” If that is so, one may question whether that is strongly probative of the proposition that disclosure would have an impact on the honesty of such donors.

The same comments apply to the question of whether the donors would donate again. Apparently, 18% changed their answer from “strongly agreed” to simply “agreed,” and the number who responded either “not sure” or “not willing to donate” increased from 4% to 10%. However, Dr. Banks does not state whether this latter increase is simply in the category of people who are “not sure,” whether it is in the category of people who stated they would not be willing to donate blood again, or, if both categories increased, where the increase was. Thus, although the Court may accept Dr. Banks’s conclusions that the results of this survey show a statistically significant change in attitudes and intentions, the Court is not convinced that it is highly probative of the ultimate fact at issue.

Ultimately, because it is the Red Cross which seeks to withhold information relevant to plaintiffs’ claims, the Red Cross must shoulder the burden of demonstrating that its concerns about the safety and adequacy of the nation’s blood supply are likely to be correct. That is a very difficult factual matter to address. It might well be that, after a full trial where experts have testified on both sides, a finder of fact would be in a position to assess the likelihood that the Red Cross’ position is the correct one. The affidavits submitted by the Red Cross, although they are competent pieces of evidence, are simply not sufficiently probative of the proposition to permit this Court to conclude, in the context of a case where the donor is dead and the Court will take great care to insure that any invasion into the private affairs of members of the donor’s family will be as minimal as possible, that permitting limiting disclosure will have a serious or significant impact on the blood supply either of the Central Ohio region or the nation generally. Consequently, although the Court is inclined to give this argument some weight simply because of the potential that great harm will occur if the adequacy or safety of the blood supply is affected, the absence of strongly probative evidence of this proposition suggests, at a minimum, that this argument not be found to be dispositive.

D. Balancing the Interests.

On the basis of the record in this case, the Court has concluded that the plaintiffs have a relatively strong interest in pursuing the discovery in question, that the residuum of privacy interests which may have survived the donor’s death can be appropriately addressed by a protective order, and that the Red Cross’ interests cannot be assigned dispositive weight. Those findings lead directly to the conclusion that further inquiry into the identity and circumstances of this blood donor should be permitted. Essentially, the Court has concluded that it is reasonably certain that the plaintiffs will be handicapped in their efforts fully to present their negligence claims if no further discovery on this issue is permitted; that the common law of privacy strongly supports the conclusion that a deceased person’s privacy interests are not to be accorded a significant amount of weight; and that the Red Cross has not produced sufficient probative evidence in support of its concern that, by permitting limited discovery in this case, the safety and adequacy of the nation’s blood supply will be affected.

It is important to stress, however, that plaintiffs have succeeded so far in making only a threshold showing that some additional discovery may be permissible under Rule 26. In order to guard against the possibility that unwarranted public disclosures of sensitive information are made, in the context where the public could easily misinterpret the circumstances under which the Red Cross’ assurance of confidentiality to blood donors may prove ineffective, and in order to protect any persons from whom discovery might be sought from annoyance or embarrassment, the Court intends to supervise each step of the discovery process and to make sure that counsel for both parties have a full opportunity for input on what the parameters of such discovery will be. The Court also notes that violations of any provisions of these orders could well lead to sanctions being imposed which could include, inter alia, dismissal of this action against the Red Cross, which was the remedy chosen by the Coleman court.

Although the reasons for the Court’s conclusions in this case should be clear from the preceding discussion of the nature of the competing interests involved, a comment on why this Court has reached a different result from the Coleman court may be appropriate. The different result in Coleman, after all, was affirmed by the Court of Appeals for the Sixth Circuit, and the decisions of that court are binding on this Court.

The first distinguishing factor, of course, is that Coleman did not deal with a donor who had died. Second, it appears that in Coleman, the donor denied being a member of any high-risk group, thus making it less likely that further inquiries to the donor or to others would have produced the type of information which would help plaintiffs’ case. Third, the Coleman court accepted a different view of the risk to the safety and adequacy of the nation’s blood supply based upon the factual record in that case. See Coleman v. American Red Cross, 130 F.R.D. 360 (E.D.Mich.1990). This Court is free to reach a different factual conclusion based upon the record before it.

The Court of Appeals’ reviewed the result reached in Coleman under an abuse of discretion standard, and the District Court's factual findings were reviewed under a clearly erroneous standard. The Sixth Circuit did not hold that a District Court, faced with a different factual record, could not make different factual findings or enter some other order with respect to discovery, but only that the Coleman court had not abused its discretion nor reached factual conclusions which were clearly erroneous. Coleman v. American Red Cross, 979 F.2d 1135 (6th Cir.1992). In this Court’s view, the Sixth Circuit’s affirmance left open the possibility that a different District Court, faced with a different record, could reach entirely different conclusions and still not have abused its discretion in dealing with this admittedly sensitive issue.

As a final matter, this and other courts should constantly keep in mind the maxim that “hard cases make bad law.” This is a hard case not because the question of whether a protective order should issue is one which the courts rarely deal with, or one which involves unfamiliar decision-making principles, but because of the current public concern over AIDS, and the need to consider whether the Court’s decision might have an impact on a matter of such great public concern as the safety and adequacy of the nation’s blood supply. The Court must strongly resist the temptation to consider these two issues to be of such significance that they automatically overwhelm the Court’s duty to permit parties to a lawsuit to have a full and adequate opportunity, through discovery, to learn facts crucial to the theories of recovery or theories of defense. At the same time, however, the Court must insure, following traditional judicial decision-making standards, that it has accorded each party’s interest the appropriate weight. This Court has attempted to do so through a careful analysis of the facts of record and of the legal protections afforded to the parties’ competing interests. These are undoubtedly matters about which other eminently reasonable jurists could reach diametrically different conclusions. Nevertheless, it is the considered judgment of this Court that the result reached in this Opinion and Order is the appropriate one, and that the plaintiffs’ motion to compel discovery should be conditionally granted.

II.

Based upon the foregoing, it is ordered that the motion of plaintiffs to compel discovery as to the identity of the donor in question is conditionally granted. No further discovery on that issue shall proceed, however, until the Court makes further orders in that regard. The Clerk is directed to contact counsel for the parties to establish a time for the holding of a discovery conference at their earliest convenience. Counsel shall be prepared to discuss, at that conference, the parameters under which further discovery on this issue should proceed.

Any party may, within ten (10) days after this Order is filed, file and serve on the opposing party a motion for reconsideration by a District Judge. 28 U.S.C. Section 636(b)(1)(A), Rule 72(a), Fed.R.Civ.P.; Eastern Division Order No. 91-3, pt. I., F., 5. The motion must specifically designate the order or part in question and the basis for any objection. Responses to objections are due ten days after objections are filed and replies by the objecting party are due seven days thereafter. The District Judge, upon consideration of the motion, shall set aside any part of this Order found to be clearly erroneous or contrary to law.

This order is in full force and effect, notwithstanding the filing of any objections, unless stayed by the Magistrate Judge or District Judge. S.D.Ohio L.R. 72.4.