Walls v. Armour Pharmaceutical Co.

*1469 OPINION RE ARMOUR’S RULE 50 MOTION FOR JUDGMENT AS A MATTER OF LAW TABLE OF CONTENTS Page I. Standard for Rule 50 Motions............................................1470 II. Armour's Statute-of-Limitation Defense....................................1470 A. The Applicable Statute of Limitation...................................1472 B. Application of the Products Liability Statute of Limitation...............1474 1. Relevant Case Law................................................1474 2. Discussion.........................................................1476 III. Armour’s “Learned Intermediary” Defense to Proximate Cause..............1481 A. Relevant Law........................................................1482 1. Federal Law Applicable to Blood Products...........................1482 2. Florida Case Law on the “Learned Intermediary” Doctrine...........1483 B. The Relationship of the “Learned Intermediary” Doctrine and the Nature of Plaintiffs Claim....................................................1484 C. Evidence Regarding Prescribing Physician’s Alleged Independent Knowledge of the Risk...................................................1487 D. Discussion............................................................1491 TV. Plaintiffs “But For” Evidence and “Cause-in-Fact” Evidence................1494 A. Relevant Case Law...................................................1494 B. Evidence.............................................................1496 C. Discussion............................................................1501 V. Conclusion ..............................................................1504 OPINION HILLMAN, Senior District Judge, Sitting by Designation. This is a wrongful death diversity action brought by Brenda Mills Walls as co-personal representative of the Estate of Jason Christopher, deceased. Brenda Mills Walls is the natural mother of Jason Christopher (“Jason”), who died on February 2, 1992. Defendant is Armour Pharmaceutical Company (“Armour”), a Delaware corporation doing business in the State of Florida. During the years 1982-1985, among other products, Armour manufactured and sold under various brand names a plasma product generically known as Factor VIII concentrate. During his entire life, Jason suffered from classic hemophilia, Type A, a hereditary bleeding disorder. Treatment required transfusions of Factor VIII concentrate. It is claimed that between January 30, 1983, and May 24, 1985, Jason used and consumed Factor VIII concentrate manufactured by Armour. As a result, plaintiff claims that Jason was infected with the Human Immunodeficiency Virus (HIV), which developed into the condition known as Acquired Immune Deficiency Syndrome (AIDS). As a result of complications caused by AIDS, Jason died on February 2, 1992, at the age of eleven. As filed on December 27, 1989, this action was originally brought by Steven R. Christopher, Jason’s father, on Jason’s behalf as a personal injury action. While this action was pending, the child died. Under Florida law, Jason’s personal injury claims were extinguished by his death. Fla.Stat. § 768.20. On April 29, 1992, Jason’s mother, Brenda Mills Walls, on behalf of the estate, filed an amended complaint for damages and demand for jury trial. The amended complaint reflected Jason’s death and asserted a wrongful death action under the Florida Wrongful Death Act, Fla.Stat. § 768.16-27. Following a six-day trial, the jury awarded total damages of $2,007,256.13. In response to special interrogatories, the jury unanimously found, from the greater weight of the evidence, the following facts: 1) that Jason Christopher was infected with the AIDS virus from Factor VIII concentrate produced and sold by Armour Pharmaceutical Corporation (“Armour”), Verdict, Question #1; 2) that Armour was negligent in failing to warn prescribing physicians in a timely or an effective manner of a potential AIDS risk associated with its Factor VIII concentrate product, Verdict, Question # 2; and 3) that Arm *1470 our’s negligence was a proximate cause of Jason Christopher’s death, Verdict, Question # 3. The jury awarded damages of $1 million to Brenda Mills Walls, Jason’s mother; damages of $1 million to Steven R. Christopher, Jason’s father; and damages to Jason’s estate of $7,256.13, for funeral expenses, Verdict, Question #4). In addition, the jury found, from the greater weight of the evidence, that Jason’s parents, Steven R. Christopher and Brenda Mills Walls, did not know or should not have known before December 27, 1985, 1) that Jason was infected with the AIDS virus, Verdict, Question # 5; or 2) that there was a potential causal connection between Jason’s HIV-infection and his use of Factor VIII concentrate, Verdict, Question #6. Presently before the court is Armour’s timely renewed Fed.R.Civ.P. 50 motion for judgment as a matter of law. In the alternative, Armour has also filed a motion for a new trial pursuant to Fed.R.Civ.P. 59. At the close of plaintiffs case on January 15, 1993, Armour presented a motion for judgment as a matter of law pursuant to R. 50(a) on the following grounds: first, that the action was time-barred, and second, that plaintiff had failed to present legally sufficient evidence as to either proximate cause or causation in fact. The court denied Armour’s motion in a bench ruling on January 19,1993. Armour renewed its motion at the close of all the evidence on January 20, 1993, pursuant to Rule 50(b). The court again denied Armour’s motion, and submitted the ease to the jury. Armour here renews its motion for judgment as a matter of law. I. Standard for Rule 50 Motions Under Rule 50, a court should grant judgment as a matter of law “if, under the governing law, there can be but one reasonable conclusion as to the verdict.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 250, 106 S.Ct. 2505, 2511, 91 L.Ed.2d 202 (1986). The Eleventh Circuit has set forth the standard for district courts to apply in ruling upon Rule 50 motions as follows. District courts should “consider all of the evidence — not just that evidence which supports the non-mover’s case — but in the light and with all reasonable inferences most favorable to the party opposed to the motion.” Verbraeken v. Westinghouse Elec. Corp., 881 F.2d 1041, 1045 (11th Cir.1989) (quoting Boeing Co. v. Shipman, 411 F.2d 365, 374 (5th Cir.1969) (en banc)), cert. dismissed, 493 U.S. 1064, 110 S.Ct. 884, 107 L.Ed.2d 1012 (1990). If the facts and inferences “point so strongly and overwhelmingly in favor of one party” that the court believes that reasonable people “could not arrive at a contrary verdict,” it is proper to grant such motions. However, if there is “substantial evidence” opposed to the motions, that is, “evidence of such quality and weight that reasonable and fairminded” people “in the exercise of impartial judgment might reach different conclusions,” such motions should be denied. Verbraeken, 881 F.2d at 1045 (quoting Boeing, 411 F.2d at 375). On the other hand, determining credibility remains “a matter solely for the jury.” Therrell v. Georgia Marble Holdings Corp., 960 F.2d 1555, 1567 (11th Cir.1992). “ ‘[I]t is the function of the jury as the traditional finder of the facts, and not the Court, to weigh conflicting evidence and inferences, and determine the credibility of witnesses.’ ” Id. (quoting Boeing, 411 F.2d at 375). II. Armour’s Statute-of-Limitation Defense Florida has for many years recognized a products liability cause of action called “negligent failure to warn.” Tampa Drug Co. v. Wait, 103 So.2d 603 (Fla.1958). The distinction between this cause of action and one for strict liability against products manufacturers has repeatedly been recognized by Florida’s courts. See, e.g., Stanley Industries, Inc., v. W.M. Barr & Co., 784 F.Supp. 1570 (S.D.Fla.1992); High v. Westinghouse Electric Corp., 610 So.2d 1259 (Fla.1992); Advance Chemical Co. v. Harter, 478 So.2d 444 (Fla. 1st DCA 1985), review denied, 488 So.2d 829 (Fla.1986); Moorman v. American Safety Equipment, 594 So.2d 795 (Fla. 4th DCA 1992), review denied, 606 So.2d 1164 (Fla.1992). 1 *1471 However, Armour claims that the applicable statute of limitation in this action is not Florida’s four-year products liability statute of limitation, Fla.Stat. § 95.11(3)(e), but is instead Florida’s four-year negligence statute of limitation, Fla.Stat. § 95.11(3)(a). Armour reaches this conclusion because it claims that Florida’s “blood-shield” statute, Fla.Stat. § 672.316(5), precludes strict products liability and warranty actions relating to the processing and distribution of blood derivatives. If Florida’s negligence statute of limitation applied to this action, Armour claims that the action is time-barred. In a bench ruling on January 19, 1993, the court ruled that Florida’s four-year products liability statute of limitation applied to the failure-to-warn action concerning Armour’s Factor VIII concentrate filed on Jason’s behalf by his father. In addition, the court ruled that this cause of action was not time-barred. Armour claims that these rulings were in error. Armour further maintains that this action is time-barred even if the products liability statute of limitation applied. This action has been complicated by,Jason’s death which occurred after his personal injury suit was filed. It is undisputed that Jason’s mother, Brenda Mills Walls, amended the complaint to reflect her son’s death and converted the claim into a wrongful death action within three months of his death. Florida’s wrongful death statute of limitation allows actions to be brought within two years from the date of the decedent’s death. Fla.Stat. § 95.11(4)(d). ' Plaintiff argues that this action is not time-barred because Mrs. Walls brought the present wrongful death action on behalf of Jason’s estate well within the two-year limitation period. I conclude that Florida’s wrongful death statute of limitation applies to the wrongful death action presently before the court, and that Mrs. Walls brought this wrongful death action well within the statute of limitation for such actions. However, in order to decide whether a wrongful death cause of action survived Jason’s death, the court must determine whether the applicable statute of limitation had run with respect to the personal injury claim originally brought on behalf of Jason. For the reasons discussed below, in order to decide this question, I must decide whether the Florida negligence or products liability statute of limitation applied to the personal injury claim originally filed on Jason’s behalf. Previous rulings of the Supreme Court of Florida have made clear that, if at the moment of death the decedent had no claim, then his survivors have no claim that can be brought in a later wrongful death action. See Nissan Motor Co. v. Phlieger, 508 So.2d 713 (Fla.1987); Variety Children’s Hospital v. Perkins, 445 So.2d 1010 (Fla.1983); Hudson v. Keene Corp., 445 So.2d 1151 (Fla. 1st DCA 1984), approved, 472 So.2d 1142 (Fla.1985). When the Phlieger decedent died, the applicable statute of limitation on his own claim had not yet run. Phlieger, 508 So.2d at 715. In allowing plaintiff in Phlieger to bring a wrongful death action, the Supreme Court of Florida distinguished the facts at issue in that case from the facts at issue in its earlier Perkins. Phlieger, 508 So.2d at 715. In Perkins, where the decedent had filed a personal injury action and obtained full recovery against the tortfeasor, the court held that a subsequent wrongful death action was barred. Id. (quoting Perkins, 445 So.2d at 1012). Further, in Hudson, the statute of limitation for the decedent’s prior cause of action had run during his lifetime, and the decedent himself would have been barred from recovery. In this situation, the court held that a subsequent wrongful death action based upon the same tortious conduct was barred. Hudson, 445 So.2d at 1153. I therefore conclude that plaintiffs wrongful death action would be time-barred if the applicable statute of limi *1472 tation had run on Jason’s personal injury cause of action before his father filed suit on his behalf on December 27, 1989. 2 Armour argues that the statute of limitation on Jason’s personal injury claim began to run in May 1985, when certain information appeared in his medical records. Armour claims that constructive knowledge of this information should be imputed to Jason’s parents at that time, regardless of whether the products liability or negligence statute of limitation applied to the original personal injury action filed on Jason’s behalf by his father. If the statute of limitation on Jason’s claim began to run in May 1985, this action would be time-barred. Jason’s father filed the original personal injury action on Jason’s behalf on December 27, 1989, more than four years later than May 1985. Further, in claiming that the negligence statute of limitation applied to this action, Armour relies upon Florida case law stating that a negligence cause of action accrues when a plaintiff has notice of either the negligence or the injury. Before the statute of limitation begins to run in a products liability ease, however, Florida courts have stated that a plaintiff must not only have notice of either the negligence or the injury, but must also know that the connection between the negligence and the injury is possibly causal. In order to decide whether the statute of limitation on Jason’s own cause of action had run before he died, I must therefore decide whether the products liability or negligence statute of limitation applied to Jason’s personal injury cause of action. A. The Applicable Statute of Limitation Relying primarily upon Silva v. Southwest Florida Blood Bank, Inc., 601 So.2d 1184, 1188 (Fla.1992), Armour claims that Florida’s “blood-shield” statute, Fla. Stat. § 672.316(5), precludes strict products liability and warranty actions relating to the processing and distribution of blood derivatives. This portion of the statute is given under the heading “Exclusion or modification of warranties.” It provides the following: The procurement, processing, storage, distribution, or use of whole blood, plasma, blood products and blood derivatives for the purpose of injecting or transfusing the same, or any of them, into the human body for any purpose whatsoever is declared to be the rendering of a service by any person participating therein and does not constitute a sale, whether or not any consideration is given therefor; and the implied warranties of merchantability and fitness for a particular purpose are not applicable as to a defect that cannot be detected or removed by a reasonable use of scientific procedures or techniques. Fla.Stat. § 672.316(5). Armour further cites Silva for the proposition that the Florida Supreme Court has stated that the statute of limitation applicable to negligence claims regarding blood derivatives is the statute governing pure negligence actions. After a careful review of the Silva case, I conclude that the Supreme Court of Florida’s narrow holding in this case does not support the broad claims that Armour maintains in the case presently before me. The narrow issue presented to the Supreme Court of Florida in Silva was whether blood banks were subject to the two-year medical malpractice statute of limitations or the four-year negligence statute of limitation. Because the court decided that blood banks were not health care providers, it concluded that blood banks were not subject to the medical malpractice statute of limitations, but were instead subject to the negligence statute of limitation. Silva, 601 So.2d at 1189. Silva cannot be read as ruling on any issues relating in any way to either pharmaceutical manufacturers or products liability actions. Certainly, Silva specifically cannot be read as barring application of the products liability statute of limitation in an action alleging failure to warn against a pharmaceutical manufacturer. The Supreme Court of Florida made no such rulings in Silva because such issues were never before the Silva court. However, the Silva court emphasized that chapter 69-157 had “nothing to do with ei *1473 ther medical malpractice or the statute of limitations. Rather, it created a ‘blood shield’ statute within Florida’s Uniform Commercial Code for the purpose of eliminating actions for strict liability against blood banks.” Silva, 601 So.2d at 1188. The court further stated that § 672.316(5) “was enacted to limit the Uniform Commercial Code [“U.C.C.”] warranties in the context of the sale of blood by declaring such a sale to be a ‘service’.” Id. Moreover, the court emphasized that “[t]here is no. evidence that the legislature intended this legal fiction (that selling blood is a ‘service’ rather than a ‘sale’) to apply in any other context.” Id. In discussing the legislative purpose of Florida’s “blood-shield” statute, the Supreme Court of Florida in Silva thus stated unequivocally that § 672.316(5) was enacted to eliminate actions for strict liability against blood banks and to limit U.C.C. warranties in the context of the sale of blood by blood banks. The Florida legislature undoubtedly had specific public policy reasons for taking these actions to protect blood banks that were supplying blood at a time when they presumably had no way of knowing whether the blood contained a defect. The Supreme Court of Florida clearly did not state that § 672.316(5) was also enacted, as Armour maintains, to convert failure-to-warn products liability claims against pharmaceutical manufacturers into ordinary negligence claims. As I noted in my January 19, 1993, bench ruling on this issue, to uphold Armour’s claim that § 672.316(5) eliminated such products liability claims would be applying the “legal fiction” that selling blood is a “service” rather than a “sale” in a totally unrelated con-To do so would be to act contrary to the legislative purpose of § 672.316(5), as specifically stated by the Supreme Court of Florida in Silva. This I decline to do. Neither the plain meaning of the statutory language of Florida’s “blood-shield” statute, § 672.316(5), nor the Supreme Court of Florida’s discussion in Silva of this statute’s legislative purpose lends any credence to Armour’s position that § 672.316(5) limits failure-to-warn products liability claims against pharmaceutical companies. Moreover, the other cases that Armour cites do not support this position. 3 text. The Florida appellate court’s discussion of the statutory purpose of § 672.316(5) in Sicuranza v. Northwest Florida Blood Center, Inc., 582 So.2d 54, 55-56 (Fla. 1st DCA 1991), further undercuts Armour’s position that Florida’s “blood-shield” statute limits failure-to-warn products liability claims against pharmaceutical companies. In Sicuranza, plaintiff brought an action against a blood bank for supplying her with HIV-infected blood. The trial court granted the blood bank’s motion for summary judgment. The appellate court reversed, holding that plaintiff did not have to prove that the defect in the blood was detectable or removable by reasonable scientific procedures or techniques to recover in tort. The Sicuranza court based its holding on its statutory interpretation that § 672.316(5) “does not, on its face, apply to tort actions such as negligence” (emphasis added). Instead, by its own terms, § 672.316(5) pertains “only to allegations of breach of implied warranties of fitness and merchantability” (emphasis added). *1474 Sicuranza, 582 So.2d at 55-56. 4 In the case presently before me, plaintiff brought a failure-to-warn products liability claim, not a claim for breach of an implied warranty of fitness or merchantability. I conclude that Fla.Stat. § 672.316(5), Florida’s “blood-shield” statute, has not in any way limited plaintiffs ability to bring such a claim nor converted such a claim into a pure negligence action. I therefore conclude that the products liability statute of limitation, not the negligence statute of limitation, applied to the personal injury action filed on Jason’s behalf on December 27, 1989, by his father. 5 B. Application of the Products Liability Statute of Limitation 1. Relevant Case Law Having concluded above that this case is governed by the products liability statute of limitations, the court must now decide whether this action was timely filed under that statute. Florida appellate courts have recently outlined the factors that must exist before Florida’s products liability statute of limitation is triggered. Babush v. American Home Products Corp., 589 So.2d 1379, 1381 (Fla. 4th DCA 1991), review denied, 599 So.2d 1278 (Fla.1992) (citing University of Miami v. Bogorff, 583 So.2d 1000 (Fla.1991)). In a case involving a prescription drug, the Babush court stated that the following factors bear on the issue as to when the facts giving rise to the cause of action were or should have been discovered: “(1) awareness of the existence of a serious physical injury; (2) knowledge that the particular drug had been administered; and (3) constructive knowledge of medical opinion in the hospital records that the drug may have contributed to the injury.” Babush, 589 So.2d at 1381 (citing Bogorff, 583 So.2d 1000). In addition, the Supreme Court of Florida, in a products liability context, also stated that plaintiffs need to have notice of the “possible invasion of their legal rights.” Babush, 589 So.2d at 1381 (citing Bogorff, 583 So.2d at 1004). 6 In Babush, the Florida appellate court interpreted the above test that the Supreme Court of Florida had fashioned in Bogorff “as having two essential ingredients: an injury distinct in some way from conditions naturally to be expected from the plaintiffs condition, and (as opposed to or in the medical malpractice context) exposure to the product in question.” Id. The court emphasized *1475 that “Use of the conjunction ‘and’ in this equation necessarily implies that the connection must be to some extent causal. There could be no ‘invasion of their legal rights’ unless this were so.” Id. (emphasis added). 7 The facts of the Babush and Bogorff cases demonstrate what sort of knowledge Florida courts have required plaintiffs in products liability cases to have before the statute of limitation is triggered. In Babush, plaintiff was diagnosed as having fatal liver damage two months after he stopped taking the drug in question. Eleven years later, plaintiff filed suit. The circuit court held that the action was time-barred and granted summary judgment to defendant drug distributor. The appellate court reversed and remanded, holding that an issue of fact existed as to when plaintiff knew or should have known that using the drug might have caused liver disease. The appellate court found “nothing in the record to show that Babush knew, earlier or from any other source, or should have known (by virtue of his own medical records) that there was a causal connection between the drug and the injury.” Babush, 589 So.2d at 1381-82. In Bogorff, minor plaintiff became comatose and then completely disabled three months after he stopped taking the drug in question. Ten years later, plaintiffs parents filed suit. The circuit court held that the action was time-barred and granted summary judgment to defendant drug manufacturer. The appellate court reversed and remanded. The Supreme Court of Florida quashed the appellate court’s decision because ten years earlier (1) the parents were “clearly aware” of their child’s “paralyzed and brain-damaged” condition and (2) they also knew that the child had been treated with the drug. The court stated that such knowledge, while not rising to the level of legal certainty, was sufficient to give notice of the “possible invasion of their legal rights” had the parents exercised reasonable diligence: “This is not a case where a drug was ingested and the alleged effects did not manifest themselves until years later.” Bogorff, 583 So.2d at 1004. In addition, both the Babush and the Bogorff courts, following Nardone v. Reynolds, 333 So.2d 25 (Fla.1976), imputed to plaintiffs knowledge of information contained in their medical records. In Nardone, the Supreme Court of Florida answered questions certified to it by the Fifth Circuit Court of Appeals as to when the statute of limitations began to run in a medical malpractice action. The Supreme Court of Florida held that, under the “peculiar facts” of Nardone (minor plaintiff became comatose and then totally blind and irreversibly brain-damaged after an operation), “knowledge of the medical, doctor, hospital, etc. records concerning the incompetent minor patient which are of a character as to be obtainable by, or available to, the patient but the contents of which are not known should be imputed to the parents, etc.” Nardone, 333 So.2d at 34. In Babush, the court applied the Nardone rule of imputed knowledge and found nothing in plaintiffs medical records that would cause his suit to be time-barred. Specifically, the court found nothing in these records to show that before 1986 (two years before plaintiff filed suit and nine years after he stopped taking the drug in question) plaintiff either knew or “should have known (by virtue of his own medical records) that there was a causal connection between the drug and the injury.” Babush, 589 So.2d at 1382. On the other hand, in Bogorff the Supreme Court of Florida concluded that plaintiffs parents had both actual and constructive knowledge sufficient to trigger running of the products liability statute of limitation more than four years before they filed suit. First, the court concluded that plaintiffs parents had sufficient actual knowledge of possible causation in 1972 because they knew that (1) the drug was administered and (2) their child subsequently suffered a “dramatic” de *1476 terioration in his physical condition. 8 Second, and in the alternative, the court applied the Nardone rule of constructive knowledge and concluded that plaintiffs parents had sufficient constructive knowledge of possible causation because medical opinion that the drug may have contributed to the child’s injury was available in the child’s medical records in 1977. In either event, the court concluded that the parents had sufficient knowledge to begin running of the limitation period more than four years before they filed a products liability claim against the pharmaceutical manufacturer in 1982. The court therefore held that the action was time-barred. 2. Discussion Because of the Nardone rule that knowledge of medical records available to a minor patient should be imputed to the parents, Armour argues that the statute of limitation on Jason’s personal injury claim began to run in May 1985, when certain information appeared in his medical records. If the statute of limitation on Jason’s claim began to run in May 1985, this action would be time-barred. Jason’s father filed the original personal injury action on Jason’s behalf on December 27, 1989, more than four years later than May 1985. The information incorporated into Jason’s medical records in May 1985 includes a hospital discharge summary and two laboratory reports. Def. Ex. 52A, 52B, 52C. The hospital discharge summary is dated 5/22/85 and is signed by Jason’s treating physician, Jerry L. Barbosa, M.D. Def. Ex. 52A. In this document, Dr. Barbosa summarized Jason’s condition while Jason was being treated in the hospital for a tooth abscess. Dr. Barbosa reported the numerical results of a lymphocyte enumeration obtained on Jason at that time. Jason “was noted to be lymphpenic with a total lymphocyte count of 828/ cu mm. He had 86% T-lymphocytes and 17% B-lymphocytes. T4 17% and T8 64% with a T4 to T8 ratio of 0.3.” Def. Ex. 52A. Dr. Barbosa further stated that “Blood was also obtained for HTLV-3. The results are pending at the time of this dictation.” Def. Ex. 52A. A laboratory report then gives as follows the results of the “miscellaneous lab test” ordered: TEST INTERPRETATION HTLV-III POSITIVE. Def. Ex. 52C. Another laboratory report, a “clinical immunology lab report” dated 5/24/85, gave a more detañed breakdown of the results that Dr. Barbosa had listed in his discharge summary. Def. Ex. 52B. This document included lines reporting the results of the T4 T-Lymphocyte subset as 17%, 141/cu mm, with ref. range of 38-52% of lymphocyte; of the T-8 T-Lymphocyte subset as 64%, 530/cu mm, with ref. range of 22-76% of lymphocyte; and of the T4/T8 ratio as 0.3, with ref. range of 1.0-2.2. Def. Ex. 52B. The back of this document gave the following information: T4 (helper/inducer subset): Lymphocytes with this phenotype are usually a subset of helper cells shown in vitro to cooperate with B-cells for antibody synthesis. They can also induce the generation of suppressor cells. Depletion of this subset is seen in numerous infections and immunodeficiencies. In Acquired Immunodeficiency Syndrome (AIDS), a profound decrease in absolute number of T4 phenotype-positive cells is common. T8 (suppressor/cytotoxic subset): Lymphocytes of this phenotype usually have a suppressor/cytotoxic function. They are thought to be involved in the inhibition of antibody production either by directly suppressing helper cells or directly turning off B-cell differentiation. An absolute or a relative increase of T8 positive cells is seen in acute mononucleosis, eytomegalo-virus infection, and acute and chrome hepatitis type B. Several reports have shown decreased numbers of T8 positive cells in rheumatoid arthritis and BLE. T4/T8 (helper/suppressor ratio): The reversal of this ratio is an indicator of altered immunoregulatory function. Changes in this ratio are seen in numerous conditions and are not diagnostic of *1477 any particular problem. Caution is necessary in interpreting this phenomenon, and special attention must be given to the absolute number of T4 + and T8 + cells. Very low absolute numbers of T4 + cells result in an inverted T4/T8 ratio and are seen in over 95% of AIDS patients with opportunistic infections and/or Kaposi sarcoma. On the other hand, increased number of T8 + cells resulting in a reversed T4/T8 ratio is often seen in apparently healthy homosexuals. Def. Ex. 52B. It was undisputed at trial that Dr. Barbosa, Jason’s treating physician, did not share any of the above information with Jason’s parents in May 1985. Both of Jason’s parents testified at trial that neither of them learned or had any suspicion that Jason was infected with HIV until 1988. Based upon my observing them as witnesses, I conclude, as the jury did, that both parents’ testimony in this regard was credible. Considering all the testimony at trial, and specifically the jury’s response to special interrogatories # 5 and # 6, in the light and with all reasonable inferences most favorable to plaintiff, as I must, I conclude that Steven R. Christopher and Brenda Mills Walls did not know before December 27, 1985, 1) that Jason was infected with the AIDS virus; or 2) that there was a potential connection between Jason’s HIV-infection and his use of Factor VIII concentrate. I therefore conclude that Jason’s parents had no actual knowledge before December 27, 1985, that would cause this action to be time-barred. 9 Armour nevertheless argues that the existence of the above information in Jason’s medical records was sufficient to give his parents constructive notice in May 1985 that Jason was infected with HIV, the virus that causes AIDS. Armour further argues that this constructive notice was sufficient to begin running of the statute of limitation in May 1985. I have carefully reviewed the Florida case law cited above and that cited by the parties, particularly Armour, 10 in their briefs on this *1478 issue. As discussed above, before the Supreme Court of Florida’s recent decision in Tanner, knowledge of an injury alone — without the additional knowledge that the injury had possibly been caused by negligence— could have been enough to trigger running of Florida’s medical malpractice statute of limitations. See Bogorff, 583 So.2d at 1002. 11 However, to trigger running of Florida’s products liability statute of limitation, Florida courts had required that plaintiffs have knowledge beyond that of the injury alone. Specifically, Florida courts had required that products liability plaintiffs have knowledge that the connection between the injury and use of the product in question was “to some extent causal.” Babush, 589 So.2d at 1381 (citing Bogorff, 583 So.2d 1000). Thus, before Tanner, Florida courts considered that products liability plaintiffs, but not medical malpractice plaintiffs, needed knowledge of possible causation in order to trigger running of the limitation period. I reiterate this distinction because Armour relies upon several medical malpractice cases decided before Tanner in support of its position that this action is time-barred under either the products liability or the negligence statute of limitation. For the record, I note that it is the two-year medical malpractice statute of limitations (given in Fla.Stat. § 95.11(4)(b)) that applies to medical malpractice actions. It is neither the products liability nor the negligence statute of limitation at issue above (given in Fla.Stat. § 95.-ll(3)(e) and 95.11(3)(a), respectively). Moreover, as discussed above, the Supreme Court of Florida held in Silva that blood banks were subject to the negligence statute of limitation, not the medical malpractice statute of limitations, because blood banks were not health-care providers. Silva, 601 So.2d at 1189. Since pharmaceutical companies are also not health-care providers, I note that an analysis analogous to the one used by the Supreme Court of Florida to decide Silva would not necessarily consider medical malpractice cases as the most appropriate authority upon which to rely in deciding either a products liability or a negligence case. Nevertheless, to the extent that Armour unrelentingly relies upon such authority, its position has been severely undercut by the Supreme Court of Florida’s recent decision in Tanner. As discussed above, in Tanner the Supreme Court of Florida clearly disapproved the rule it formerly applied in medical malpractice actions; that is, that awareness of a causal connection between the injury and the negligence was not needed to begin run *1479 ning of the statute of limitations. The Supreme Court of Florida has now changed its former rule to require that knowledge of the injury sufficient to begin running of the medical malpractice statute of limitations include knowledge of “a reasonable possibility that the injury was caused, by medical malpractice.” Id. 618 So.2d at 181 (emphasis added). However, even before Tanner, the knowledge of the injuries that courts in both medical malpractice and products liability cases considered sufficient to trigger running of the statutes of limitations was clearly more obvious and definitive than the constructive knowledge of Jason’s HIV-infection that Armour argues is sufficient here. For example, the Babush plaintiff was diagnosed as having fatal liver damage two months after he stopped taking the drug in question. The Bogorff plaintiff became comatose and then completely disabled three months after he stopped taking the drug in question. Moreover, medical opinion that the drug may have contributed to the child’s injury was available in the child’s medical records more than four years before the parents filed suit. The Steiner plaintiff lost eyesight after taking a particular drug. The Nardone plaintiff became comatose and then totally blind and irreversibly brain-damaged after an operation. The Norsworthy plaintiff, hospitalized for difficulty breathing, was left with subglottic stenosis (the narrowing of the airway below the vocal cords) after treatment that included intubations and a tracheotomy. The MacMurray plaintiff knew that she had Hodgkin’s disease and, through her husband, a physician, had control of her medical records more than two years before she filed suit. These records contained evidence of an error in interpreting plaintiffs X-rays that defendant admitted constituted actionable negligence. Even in a medical malpractice context, where less “knowledge” was formerly required by Florida courts than in a products liability context, the “injuries” in all of these other cases were obvious and definitive enough in and of themselves to provide plaintiffs knowledge of a “possible invasion of their legal rights” sufficient to begin running of the statute of limitations. Such is not the case in the action presently before me. In my bench ruling of January 19, 1993, I referred to the condition which the Nardone court believed should be satisfied before a court should impute constructive knowledge to the parents of the contents of a child’s medical records. The court stated that these records should be “of a character as to be obtainable by, or available to, the patient.” Nardone, 333 So.2d at 34. 12 In Nardone, before deciding that constructive knowledge of the child’s medical records should be imputed to his parents, the court emphasized that the “nature of the child’s condition [becoming totally blind and irreversibly brain-damaged after an operation] was obvious and known to the plaintiffs.” Nardone, 333 So.2d at 34. Clearly, this is a very different type of injury than the one suffered by the child here. By May 1985 Jason had presented no clinical symptoms that his parents could recognize as, or that they had been told were, symptoms of HIV-infection. It is undisputed that by May 1985 Jason had already been injured in that he was already infected with HIV, the virus that causes AIDS. From a medical perspective, the fact of Jason’s HIV-infection may be said by that time to be “an injury distinct in some way from conditions naturally to be expected from the plaintiffs condition.” Babush, 589 So.2d at 1381. The Babush court considered such an injury to be one of the two “essential ingredients” which, when joined with the *1480 knowledge that the link between these ingredients was “to some extent causal,” was sufficient to begin running of the products liability statute of limitation. Id,. 13 However, it is also undisputed that in May 1985 the fact of Jason’s HIV-infection was discerned only by medical laboratory testing of the cellular components of his blood. Thus, Jason’s condition in May 1985 with regard to his HIV-status cannot be said to have been “obvious or known to the plaintiffs” in the same way that the Nardone court considered minor plaintiffs blindness and brain-damage that ensued after an operation to have been “obvious or known” to his parents. I have already concluded that Jason’s parents had no actual knowledge of the test results in question. Therefore, Jason’s condition in May 1985 with regard to his HIV-status could not be said to have been known by his parents in any way except by imputing to them constructive knowledge of the significance of laboratory test results reported in medical terminology. I am not convinced that the Nardone court would have intended its 1976 holding under the “peculiar facts” of that case to be applied to impute constructive knowledge to Jason’s parents of such medical records as are at issue here in an HIV-infection case. It was apparently only because the Nardone child’s condition “was obvious and known” to his parents that the Nardone court believed it appropriate to impute knowledge of the child’s medical records to them. For the purposes of a medical malpractice action, the Nardone court clearly considered that a dramatic deterioration in a child’s condition, occurring after an operation, was sufficient to put his parents on notice of the “possible invasion of their legal rights.” Nardone, 333 So.2d at 34. The Nardone court did not rule that knowledge of a child’s condition discernible only by medical laboratory testing of the cellular components of blood and reported in the medical records only in technical terminology was sufficient to put a child’s parents on notice of the “possible invasion of their legal rights” in a products liability action. However, even imputing full knowledge of all of the information contained in Def. Ex. 52A, 52B, and 52C to Jason’s parents, I conclude that this constructive knowledge was not sufficient as a matter of law to make them aware “of the existence of a serious physical injury.” As discussed above, such awareness is the first of three factors listed by the Babush court as bearing on the issue as to when the facts giving rise to a products liability cause of action were or should have been discovered. Babush, 589 So.2d at 1381 (citing Bogorff, 583 So.2d 1000). 14 In Def. Ex. 52A, Dr. Barbosa reported, without explanation, the numerical results and percentages obtained from a lymphocyte enumeration. In addition, without explaining the significance of this action, Dr. Barbosa noted that blood was “obtained for HTLV-3.” Def. Ex. 52A. One lab report merely stated that the result of the miscellaneous lab test ordered, HTLV-III, was positive. Def. Ex. 52C. The front of the other lab report gave further numerical breakdowns on the values obtained in the lymphocyte enumeration. The highly technical “INTERPRETATION/LYMPHOCYTE ENUM.” given on the back of this report contained only two references to AIDS. Both of these references are ambiguous at best, especially when considered in the overall context of the cautionary language used immediately before and after these references. While under the heading “T4” the report explained that in AIDS “a profound decrease in absolute number of T4 phenotype-positive cells is common,” it also states that “[depletion of this subset is seen in numerous infections and immunodeficiencies.” Def. Ex. 52B. Similarly, under the heading “T4/F8” the report explained that “Changes in this ratio are seen in numerous conditions and are not diagnostic of any particular problem. Cau *1481 tion is necessary in interpreting this phenomenon, and special attention must be given to the absolute numbers of T4 + and T8 + cells.” While “very low absolute numbers of T4 + cells result in an inverted T4/T8 ratio and are seen in over 95% of AIDS patients with opportunistic infections and/or Kaposi sarcoma, [o]n the other hand, increased number of T8 + cells resulting in a reversed T4/T8 ratio is often seen in apparently healthy homosexuals.” Def. Ex. 52B. As I stated in my January 19, 1993, bench ruling, I am satisfied that the information given in the documents described above was not specific enough to impute to Jason’s parents the knowledge that, as of May 1985, their child had been infected with HIV, the virus that causes AIDS. No Florida case has ruled that information beyond that contained in the medical records should be imputed to the parents of a minor plaintiff. The information given in the documents described above did not, for instance, explain the meaning of the term “HTLV-III.” It did not explain the implications of testing positive for HTLV-III. It did not make any link whatsoever between HTLV-III and HIV or AIDS. The only two references to AIDS in these entire documents were couched in medical terminology and were so hedged about with cautionary language as to be ambiguous at best, especially to lay people. As noted later, Armour repeatedly (and properly) emphasizes that considerably less medical knowledge about AIDS existed in 1983-1984 than exists now. Likewise, and perhaps even more so, knowledge about the symptoms of AIDS among the general population at that time was practically non-existent. In short, there is nothing about the information in question that would have communicated to reasonable parents in May 1985 that their child was now infected with HIV or that their child would ultimately develop AIDS. 15 I therefore conclude that constructive knowledge of Jason’s HIV-infection should not be imputed to his parents as of May 1985. Furthermore, even imputing full knowledge of all of the information contained in Def. Ex. 52A, 52B, and 52C to Jason’s parents, I conclude that this constructive knowledge did not provide “constructive knowledge of medical opinion in the hospital records that the drug may have contributed to the injury.” As discussed above, such constructive knowledge of possible causation is the third of three factors listed by the Babush court as bearing on the issue as to when the facts giving rise to a products liability cause of action were or should have been discovered. Babush, 589 So.2d at 1381 (citing Bogorff, 583 So.2d at 1000). The information contained in the medical records was certainly not specific enough to impute to Jason’s parents the knowledge that their child’s HIV-infection had possibly resulted from receiving either Factor VIII concentrate in general or Armour’s Factor VIII product in particular. There is nothing in these records that raised the questions “May HIV-infection be related to use of concentrate?” or “May HIV-infection be related to use of Armour’s product?” The medical records at issue before me are devoid of even a suggestion of possible causation. For all the reasons set forth above, I conclude that the constructive knowledge imputed under Florida law to Jason’s parents from his May 1985 medical records did not provide notice of a “possible invasion of their legal rights” sufficient to begin running of the products liability statute of limitation in May 1985. I therefore conclude that the personal injury action brought on Jason’s behalf on December 27, 1989, by his father was not time-barred. I further conclude that Brenda Mills Walls’s present wrongful death action is not time-barred. III. Armour’s “Learned Intermediary” Defense to Proximate Cause First, the jury found and the court agrees that in 1982-1983 Armour was or should have *1482 been aware from reasonable and available evidence of a potential risk of acquiring AIDS from contaminated blood or blood products. Plaintiff presented considerable expert testimony in support of her claim that Armour’s duty to warn of the AIDS risk associated with Factor VIII concentrate arose at an earlier time than Armour acted. This evidence took the form of oral testimony and of exhibits including both Armour's own internal documents and technical articles published in medical journals. The matter was contested and went to the jury. By its interrogatory verdict, the jury specifically found, from the greater weight of the evidence, that Armour was negligent in failing to warn prescribing physicians in a timely or an effective manner of a potential AIDS risk associated with its Factor VIII concentrate product. Verdict, Question #2. The jury further found, from the greater weight of the evidence, that Armour’s negligence was a proximate cause of Jason’s death. Verdict, Question # 3. Second, as an affirmative defense, Armour renews its claim that Dr. Barbosa, Jason’s treating physician, had independent knowledge of the risk of AIDS associated with Factor VIII products. At trial, Armour offered both testimony and exhibits consisting of documents from Jason’s medical records in support of this claim. According to Armour, even if Armour failed to warn of the risk of AIDS associated with Factor VIII products, its failure cannot have been a proximate cause of Jason’s HIV-infection. According to the well-recognized principle of the “learned intermediary” doctrine, if Jason’s prescribing physician had as much knowledge as Armour did about the risk of AIDS associated with the Factor VIII concentrate and prescribed the product anyway, then Armour’s failure to warn cannot have been a proximate cause of Jason’s injury because the physician already knew of the risk. Because Armour raises the “learned intermediary” doctrine as an affirmative defense, Armour bears the burden of proof on this issue. A. Relevant Law 1. Federal Law Applicable to Blood Derivatives Blood derivatives such as Factor VIII are prescription biologicals subject to federal regulation as both “biological products” and “drugs.” See Public Health Service Act, “Regulation of Biological Products,” 42 U.S.C. § 262; see also Food, Drug, and Cosmetic Act (“FDCA”), 21 U.S.C. §§ 301 et seq. During the relevant time period, the following regulation specified the format and content of prescription drug labeling: § 201.57 Specific requirements on content and format of labeling for human prescription drugs. (e) “Warnings”: Under this section heading, the labeling shall describe serious adverse reactions and potential safety hazards, limitations in use imposed by them, and steps that should be taken if they occur. The labeling shall be revised to include a warning as soon as there is reasonable evidence of an association with a drug; a causal relationship need not have been proved. 21 C.F.R. § 201.57(e) (1982). Moreover, the regulations governing Factor VIII as a prescription biologic licensed under the Public Health Services Act prohibited Armour from changing its labeling without prior FDA approval: § 601.12 Changes to be reported. (a) General. Important proposed changes in location, equipment, management and responsible personnel, or in manufacturing methods and labeling, of any product for which a license is in effect or for which an application for license is pending, shall be reported to the Director, Bureau of Biologies, by the manufacturer, and unless in case of an emergency, not less than 30 days in advance of the time such changes are intended to be made. (b) Manufacturing methods and labeling. Proposed changes in manufacturing methods and labeling may not become effective until notification of acceptance is received from the Director, Bureau of Biologies. 21 C.F.R. § 601.12 (1982). Under federal law, “labeling” is broadly defined and includes not only package inserts, but also separate communications con *1483 cerning the drug, such as “Dear Doctor” letters sent by drug manufacturers to physicians to provide information about a drug. See 42 U.S.C. § 262(g), which states that nothing contained in chapter 6A of the Public Health Service Act “shall be construed as in any way affecting, modifying, repealing, or superseding” the provisions of the FDCA; see also 21 U.S.C. § 321(m), which gives the FDCA definition of “labeling”: “all labels and other written, printed or graphic matter (1) upon any article or any of its containers or wrappers, or 2) accompanying such article”; see also Kordel v. United States, 335 U.S. 345, 349-50, 69 S.Ct. 106, 109, 93 L.Ed. 52 (1948), in which the court clarified that separate literature sent by the manufacturer constituted “labeling” under the FDCA when it supplemented or explained materials sent with the drug. An additional regulation provided the following information regarding mailing important information about drugs: Manufacturers and distributors of drugs and the Food and Drug Administration occasionally are required to mail important information about drugs to physicians and others responsible for patient care. In the public interest, such mail should be distinctive in appearance so that it will be promptly recognized and read. 21 C.F.R. § 200.5 (1982). This regulation further specified the type of envelope and print size for the FDA to use in such mailings. It provided that, when information in the letter “concerns a significant hazard to health,” the statement IMPORTANT DRUG WARNING must be printed on the face of the envelope in 36-point Gothic Type. Id. It “asked” manufacturers and distributors to “make such mailings as prescribed by this section.” Id. 2. Florida Case Law on the “Learned Intermediary’’ Doctrine The Supreme Court of Florida applied the “learned intermediary” doctrine in Felix v. Hoffman-LaRoche, Inc., 540 So.2d 102 (Fla.1989), to approve the lower court’s opinion that any inadequacy in the drug manufacturer’s warnings concerning its product Accutane could not have been the proximate cause of the birth defects and subsequent death of a child. At the outset, the court explained that it is the manufacturer’s duty to warn the physician, not the patient, of the dangerous side effects of a drug. Id. at 104 (citing Buckner v. Allergan Pharmaceuticals, Inc., 400 So.2d 820 (Fla. 5th DCA), review denied, 407 So.2d 1102 (Fla.1981)). “This is so because the prescribing physician, acting as a ‘learned intermediary’ between the manufacturer and the consumer, weighs the potential benefits against the dangers in deciding whether to recommend the drug to meet the patient’s needs.” Felix, 540 So.2d at 104 (citing Reyes v. Wyeth Laboratories, 498 F.2d 1264, 1276 (5th Cir.), cert. denied, 419 U.S. 1096, 95 S.Ct. 687, 42 L.Ed.2d 688 (1974)). In Felix, the lower court held that the drug manufacturer “could not be penalized for the failure of the doctor to impart knowledge concerning the dangers of the drug of which the doctor had been warned and was aware.” Felix, 540 So.2d at 104. The Supreme Court of Florida approved this result because the prescribing physician testified that he “fully understood” the warnings provided by the drug manufacturer and also had prior knowledge of the teratogenic propensity of the drug. The Supreme Court of Florida therefore agreed that any inadequacy in the warnings provided by the drug manufacturer could not have been the proximate cause of the child’s birth defects. Felix, 540 So.2d at 104. Cf. Zanzuri v. G.D. Searle & Co., 748 F.Supp. 1511 (S.D.Fla.1990), a ease in which the court denied defendant drug manufacturer’s motion for summary judgment. Plaintiff had alleged that the manufacturer negligently failed to disclose unfavorable information and provided inaccurately favorable information to the general public and physicians about its birth control product. In Zanzuri, the prescribing physician, who was generally informed about the risks of the product, testified that he relied heavily on the manufacturer’s literature in forming *1484 his opinion as to these risks. The court stated that the presence of misstatements in the manufacturer’s literature heightened the evidentiary threshold which the manufacturer must cross in order to show that the physician was “an intermediary sufficiently informed to interrupt the causal link of liability between the manufacturer and the plaintiff.” Zanzuri, 748 F.Supp. at 1518. B. The ■ Relationship Between the “Learned, Intermediary” Doctrine and the Nature of Plaintiff s Claim The following facts about the AIDS warning that Armour eventually provided with its Factor VIII concentrate product are not disputed. Armour submitted a request to the FDA to change its Factor VIII labeling to include an AIDS warning in late September 1983. The FDA approved this labeling change on January 25, 1984. Armour first published an AIDS warning in its Factor VIII labeling in February 1984. This warning read as follows: The possibility exists that Acquired Immune Deficiency Syndrome, AIDS, an immunological disorder with extremely serious consequences, may be transmitted by blood, blood products and blood derivatives, including clotting factors. However, the causative agent has neither been isolated nor identified. This information should be considered in determining patient care and treatment. Pl.Ex. 57, 58. Armour argues that, because plaintiff never challenged the specific language quoted above, the “learned intermediary” doctrine focuses plaintiffs claim on the question of whether Dr. Barbosa had independent knowledge of the AIDS risk conveyed in its eventual warning. That is, according to Armour, at the time Dr. Barbosa prescribed Factor VIII concentrate for Jason in the latter half of 1983, did Dr. Barbosa believe, in the absence of reading Armour’s as-yet-unpublished warning, that the “possibility existed” that AIDS could be transmitted by blood products? I am satisfied that an affirmative answer to this question is insufficient, as a matter of law, to invoke the “learned intermediary” doctrine as a defense to Armour’s liability in this matter. Armour’s argument misconceives the nature of the duty placed upon it in this respect by the comprehensive framework of federal law outlined above. This body of law, applicable to Armour’s product Factor VIII concentrate, clearly mandates that warning labeling shall describe serious adverse reactions and potential safety hazards, limitations in use imposed by them, and steps that should be taken if they occur. The labeling shall be revised to include a warning as soon as there is reasonable evidence of an association with a drug; a causal relationship need not have been proved. 21 C.F.R. § 201.57(e) (1982) (emphasis added). This language is not discretionary. While it is explicit that “a causal relationship need not have been proved,” it requires that pharmaceutical companies “shall” revise labeling “as soon as” there is “reasonable evidence” of an “association” of “serious adverse reactions and potential safety hazards” with a drug. Since pharmaceutical companies are bound by federal law to fulfill this duty, physicians are justified in relying on the literature from pharmaceutical companies to inform them “as soon as” there is such “reasonable evidence.” Moreover, physicians are also justified in assuming that, in the absence of a warning from a pharmaceutical company about a particular drug, “reasonable evidence” of an “association” of “serious adverse reactions and potential safety hazards” with that particular drug does not yet exist. Within the comprehensive framework of binding federal law applicable to such cases, the crux of the issue in a failure-to-warn case in which plaintiff claims that a warning was not given as soon as it should have been thus becomes very different than the crux of the issue in a failure-to-warn case in which an allegedly inadequate warning was given. Even apart from such a comprehensive framework of binding federal law, the mere fact of receiving a warning about a particular drug, as opposed to not receiving such a warning, is clearly significant in and of itself. This is especially so in the context of a *1485 pharmaceutical company’s attempt to use the “learned intermediary” doctrine to relieve itself of liability for an injury by showing that a physician was “an intermediary sufficiently informed to interrupt the causal link of liability between the manufacturer and the plaintiff.” Zanzuri, 748 F.Supp. at 1518. If physicians received an initial warning about a particular drug, albeit an allegedly inadequate one, it is reasonable to assume that, at a minimum, they were put on notice that they should undertake independent inquiry into the hazards of that particular drug. If physicians received no warning, they cannot be said to have been put on such notice. However, in the comprehensive scheme of federal regulation outlined above, the very fact of a physician’s receiving no warning from a pharmaceutical company about a particular drug communicates an additional and extremely significant message. This is so because the pharmaceutical company is bound to apply to the FDA for permission to warn of adverse reactions and potential hazards “as soon as” the pharmaceutical company knows of “reasonable evidence of an association with a drug” of “serious adverse reactions and potential safety hazards,” even in the absence of proof of a causal relationship. If the FDA is convinced by the pharmaceutical company’s “reasonable evidence,” it grants the company permission to issue a warning or to revise its labeling to include a warning. Therefore, when a physician receives a warning from a pharmaceutical company about the hazards of a particular product, the physician may reasonably conclude that — quite apart from any particular language which may or may not be used in the warning — the very fact of the pharmaceutical company’s now issuing the warning effectively means that “reasonable evidence” of an “association” of “serious adverse reactions and potential safety hazards” with that product now exists. Before receiving the first such warning, the physician is justified, under federal law, in assuming that such evidence does not yet exist. In such a ease as the one before the court, plaintiff claims that Armour did not issue a warning about its Factor VIII concentrate as soon as it should have. The fundamental issue arising in such a ease is very different from those arising in cases in which an allegedly inadequate warning was given. In the latter type of case, in which a warning was in fact given, the parties usually quibble about how much the prescribing physician actually knew of the risks inherent in using a particular drug as compared with what knowledge of such risks was conveyed by the particular language of the allegedly inadequate warning. 16 In such a case as the one before the *1486 court, however, the fundamental issue to be resolved is whether Armour in fact had “reasonable evidence” of an “association” of “serious adverse reactions and potential safety hazards” with its Factor VIII concentrate before it applied for permission from the FDA to warn of these hazards. See Moore v. Armour Pharmaceutical Co. [Transfer Binder July 1990-August 1991], Prod.Liab.Rep. (CCH) ¶ 12,665, 1990 WL 369571 (M.D.Fla. Aug. 27, 1990), in which the court denied defendant Armour summary judgment in a case similar to the one presently before me. In Moore, Armour had argued, first, that it had no duty to warn before the time that it did begin to use warning labels with its Factor VIII product and, second, that plaintiffs failure-to-warn claim was preempted by federal law governing the labeling of medical products. As to the first argument, the court denied Armour’s motion because the court concluded that “(viewing the evidence in plaintiffs’ favor) defendant was or should have been aware of respectable medical authority as early as July 1982 that there was a potential risk of catching AIDS from contaminated blood or blood products.” Id. at 37,968. As to the second argument, the court denied Armour’s motion because the court concluded that “the course of events in this case undercuts any claim that the state tort law relevant here is inconsistent with the federal law regulating the labelling of biological medical products.” Id. The court further concluded that no reason existed why Armour could not have applied to the FDA at an earlier date to insert an AIDS warning in its Factor VIII concentrate product. Id. 17 Under the comprehensive framework of federal law applicable to this case, plaintiffs major thrust has been that Armour in fact had knowledge of “reasonable evidence” of an “association” of an AIDS risk with Factor VIII concentrate at the time Jason was receiving this product and that Armour did not in fact apply to the FDA to issue a warning about this product as soon as it was bound to do. This was the principal issue around which the case was tried and ultimately submitted to the jury. The jury unanimously and specifically found against Armour on this issue. I am satisfied that Armour is not entitled to avail itself of the “learned intermediary” doctrine to escape liability for an injury caused by its Factor VIII concentrate merely by establishing that the prescribing physician was aware of the information provided in its eventual warning. Rather, Armour would only be entitled to avail itself of the “learned intermediary” doctrine to es *1487 cape such liability if it could also establish that the prescribing physician not only knew of the “possibility” of the AIDS risk set forth in the ultimate warning, but also had independent knowledge of “reasonable evidence” of an “association” of an AIDS risk with Factor VIII concentrate at the time that Armour should have issued a warning about this product, as required by federal regulations. This is so because the very act of Armour’s sending the first warning to physicians about its Factor VIII concentrate would have put physicians on notice that there was then “reasonable evidence” of an “association” of an AIDS risk with Factor VIII concentrate. This is obviously a much more serious warning than one that advises of a “possible” causal connection. After all, almost anything is possible! Under the comprehensive framework of federal law applicable to this case, the act of Armour’s sending that first warning — quite apart from any particular language which may or may not have been included in the warning — itself would have communicated this extremely significant message. Whenever it occurred, Armour’s first act of warning would essentially tell physicians that Armour and the FDA had at that time — and not before that time — agreed that there was “reasonable evidence” of an “association” of an AIDS risk with Factor VIII concentrate. To prevail on its “learned intermediary” defense in such a context, Armour must therefore establish that Jason’s prescribing physician had knowledge both of “reasonable evidence” of an “association” of an AIDS risk with Factor VIII concentrate and of the “possible” causal connection contained in Armour’s eventual warning. Moreover, Armour must show that the physician had such knowledge at the time that Armour should have issued a warning about its Factor VIII concentrate. Any other result would be illogical and would negate the communicative impact that a physician’s receiving a first warning from a pharmaceutical company about a particular product is designed to have. Any other result would countenance depriving Jason’s prescribing physician of the same level of knowledge about the risks of Factor VIII concentrate that he should have had, under binding federal law, if Armour had fulfilled its non-discretionary duty to warn in a timely and an effective manner. C. Evidence Regarding Prescribing Physician’s Alleged Independent Knowledge of the Risk In the time period in question, two products, cryoprecipitate and Factor VIII concentrate, were commonly used to help control bleeding in hemophiliacs. Cryoprecipitate is made by freezing blood plasma from one blood donor, thawing it, and then drawing off a precipitate rich in Factor VIII (the clotting factor that hemophiliacs lack). This product is effective in treating “moderate” hemophiliacs (such as Jason was). Since the blood of moderate hemophiliacs contains some clotting factor, moderate hemophiliacs usually begin bleeding only after physical trauma, such as that caused by a fall. However, it is not always possible to use cryoprecipitate to transfer sufficient clotting factor quickly enough to treat “severe” hemophiliacs effectively. Since the blood of severe hemophiliacs contains very little clotting factor, severe hemophiliacs can begin bleeding spontaneously. Factor VIII concentrate is made in an entirely different way than cryoprecipitate. It is a “concentrate” mixed at a factory from the frozen plasma of thousands of donors (commonly 3,000-5,000 or more donors) who donate their plasma at plasma collection centers. The clotting factor is pulled out of the large pool of combined plasma, freeze-dried, and placed in small bottles that are kept refrigerated until use. Before use, the concentrate crystals are mixed with sterile water to form a liquid that is injected into the patient’s veins. Despite the safety factor — that is, that cryoprecipitate comes from just one individual — it is nevertheless less convenient to receive cryoprecipitate than it is to receive Factor VIII concentrate. Patients must go to a hospital and he on a table while a “bag” of cryoprecipitate drips into their veins through an intravenous tube. Factor VIII concentrate, on the other hand, is easily ad *1488 ministered by injection and can be done at home. Using Factor VIII concentrate is thus a quicker and more convenient way to receive clotting factor than using cryoprecipitate. It is also easier to control the exact amount of clotting factor that patients receive with Factor VIII concentrate. In July 1979, Dr. Jerry L. Barbosa, Jason’s treating physician, became the Medical Director, Department of Pediatric Hematology/Oncology, All Children’s Hospital, St. Petersburg, Florida. He still holds this position. 18 At trial, Dr. Barbosa testified that he works seven days a week from 6-8 a.m., when he begins rounds, until 8-10 p.m. He sees patients most of the day in what is the largest such hematology practice in Florida and the fourth largest in the southeastern United States. Dr. Barbosa testified that he tries to keep himself abreast of new developments and that he reads journals in the time available. However, he does not always read them right away, nor is he always able to read each one. Each month he receives 10-15 medical journals or publications at his office. Dr. Barbosa is a treating physician; he testified that he is not involved in research. TT, Vol. 3 at 68-70. When asked how he kept informed of new products, Dr. Barbosa stated that he did so through the general medical literature, pharmaceutical company representatives, or the literature that pharmaceutical companies send physicians. TT, Vol. 3 at 71. If there is a significant hazard or side effect, Dr. Barbosa testified that “usually the pharmaceutical company takes care of immediately notifying us.” Id. In such a case, Dr. Barbosa expected to be notified about the hazard through a “Dear Doctor” letter or a package insert. According to Dr. Barbosa, the urgency of the situation dictated how the message was delivered. In an urgent situation, Dr. Barbosa expected to receive a “Dear Doctor” letter or a mailgram in an envelope containing “the sign urgent [or “stat,” the medical term for “urgent”] or important information for the practice inside, read it.” TT, Vol. 3 at 71-72. He testified that he depended heavily upon the pharmaceutical companies for such information. TT, Vol. 3 at 94. When asked what he had heard about AIDS in 1983, Dr. Barbosa testified that it was “ill-defined,” “new to us,” and involved the fields of infectious disease, immunology, and epidemiology. TT, Vol. 3 at 88. When asked what warnings of a possible AIDS risk from Factor VIII concentrate he had received from Armour by July 1983 (before first administering Armour’s product to Jason) or what knowledge of such a risk he had from any source, Dr. Barbosa did not recollect having received any warning or having any knowledge. TT, Vol. 3 at 89. When asked if he 'had taken any precautions because of AIDS, Dr. Barbosa stated that he did not recall. Id. When asked if because of AIDS he had modified the Factor VIII treatment of another of his patients, a child named Johnny Kellar, Dr. Barbosa testified that he continued to use Factor VIII concentrate that was not heat-treated with Johnny in 1983. TT, Vol. 3 at 90. 19 When asked how receiving a written warning from a concentrate manufacturer in July *1489 1983 that there was a possible risk of AIDS transmission via Factor VIII concentrate would have affected his care and treatment of Jason, Dr. Barbosa testified that he would have switched Jason to cryoprecipitate: A: Well, Jason Christopher, unlike other patients [such as Johnny Kellar] was a moderate hemophiliac. He could have been — could have benefit [sic] from cryoprecipitate. Q. Would you have changed his therapy? A. Mosy [sic] likely, yes, sir. Q. To what? A. To cryoprecipitate. Q. Why? A: When you give a vial of factor VIII for a patient, it comes from 10 to 20,000 donors. And you — those are donations that come from inmates, paid providers of blood and so on. Cryoprecipitate comes from one single person and those are volunteer donors. So I feel that cryoprecipitate would be a safer way to treat this patient. TT, Vol. 3 at 90-92. Q. We know that Jason received six injections of non-heat treated concentrate in the last half of 1983. Had you received a warning of a possible risk of AIDS transmission via concentrate from Armour prior to giving those injections, would you have given them? A. No. Q. What would you have done? A. Give cryoprecipitate. TT, Vol. 3 at 93-94. In questioning Dr. Barbosa on cross-examination about Def.Ex. 52C, the May 1985 test result that Jason’s HTLV-III test was positive (discussed above), Armour elicited the following testimony concerning the state of Dr. Barbosa’s knowledge in May 1985 about the risk of AIDS in the hemophiliac population: Q. And your interpretation of that [Jason’s May 1985 test result] was that he at some previous time had been infected with the virus? A. Don’t get me wrong in the sense, okay. By that time, as I mentioned before, the AIDS situation was a very ill-defined entity. Especially in hemophilia, we treaters were talking about AIDS especial [sic]. Later on when the facts were known, when we know the awful truth, then we know that this patient were active [sic] infected. Q. So it’s your understanding that as of May 1985, the fact that a hemophiliac had a positive HTLV-3 report didn’t mean that he was infected with the AIDS virus? A. That may be the thinking at that time, sir. I don’t recall exactly the specifies. Q. You don’t recall whether at that time it was believed that AIDS was caused by the HTLV-3 virus? A. In specific [sic] in the hemophilia population, I don’t know, sir. We were talking of exposure. Q. And hemophiliacs, is it not true, have often been different from normal populations in terms of their immune situations; is that right? A. Because of the factor VIII infusions even before the AIDS epidemic, sir. 20 Q. And a lot of hemophiliacs, perhaps, including yourself, I don’t know — a lot of hemophilia treaters were of the point of view early in the AIDS situation that what they were seeing in their patients or in hemophiliacs in general might not be the same thing that was being seen in homosexuals or IV drug abusers but perhaps with [sic] something related *1490 simply to their factor treatment or their blood products treatment? A. Thinking was going in both directions, sir. Q. And you were aware of that at that time; is that correct? A. I don’t recall. TT, Vol. 3 at 115-116. In the context of this testimony, Armour argues that in July 1983 Dr. Barbosa was “an intermediary sufficiently informed to interrupt the causal link of liability between the manufacturer and the plaintiff,” Zanzuri, 748 F.Supp. at 1518, because of evidence concerning Dr. Barbosa’s patient Johnny Kellar. This evidence included certain information contained in Johnny Kellar’s medical records, as well as Dr. Barbosa’s affirmative answer to the second question asked during the following exchange on cross-examination: Q. And discussing [in Def.Ex. 56, Dr. Barbosa’s July 21, 1983, letter to a Dr. Dean Fauber] his [Johnny Kellar’s] ever present cervical lymphadenopathy and observing, did you not, quote, “His T cell subset analysis shows elevated T suppressor cell number. This rules out the possibility of AIDS. He’s to be retested again in three months, period,” close quote. Is that not correct? A. Well, I’m glad that you brought up the subject. That gives you an idea of the confusion at that time of the entire picture of AIDS in interpreting this test. Right now we know that the opposite is right. Thank you, sir, for bringing up this. Q. Certainly. But in any event, you were testing Johnny Kellar’s T-cells because you had a concern that he might be in an AIDS situation as had other reported hemophiliacs; is that not true? A. Yes. You know, given the quote unquote “ignorance” that we were going through that period of time, because information known by some other people was not released for whatever reasons, so we have to use whatever our minds tell you we have at your disposal [sic] at that time. I wish the people that knew at that time the implications of this, would have notified us with their decision, so— TTE, Vol. 3 at 126-127 (emphasis added). The written evidence concerning Dr. Barbosa’s patient Johnny Kellar that Armour argues is sufficient to establish Dr. Barbosa’s status as a “learned intermediary” knowledgeable about the association of an AIDS risk with Factor VIII concentrate is contained in Def.Ex. 5, 6, 53, 54, and 56. The first document, Def.Ex. 5, is a “RECORD OF TREATMENT/CLINIC DICTATION” prepared by Dr. Mary R. Andriola on 5-26-83. At the end of this document, a line provides “cc: Dr. Barbosa.” In this document, Dr. Andriola discussed Johnny Kellar’s treatment history. Johnny, a two-year-old child with hemophilia, had suffered several intercranial hemorrhages. In her notes on Johnny’s 5-20-83 visit to the neurology clinic, Dr. Andriola stated, “The child currently is no longer receiving Factor 8 due to AIDS. Factor 8 will only be used symptomatically.” Def.Ex. 5. The second document, Def.Ex. 6, is a hospital discharge summary dated 10/10/83 and prepared by Dr. Barbosa. It summarized the details of Johnny’s symptoms and treatment leading up to and following an “emergency right temporoparietal craniotomy with evacuation of acute subdural hematoma and relief of right subtemporal decompression.” Def.Ex. 6. Because of his condition at that time, Johnny was admitted to the hospital on 8/18/83 and was not discharged until 9/15/83. At the beginning of the two-page discharge summary, Dr. Barbosa stated that “The patient has had several episodes of intracranial bleeding in the past for which he has been kept on prophylactic Factor VIII twice a week. The prophylactic course was discontinued a few months ago and he was receiving his Factor VIII on a p.r.n. [as-needed] basis.” Def.Ex. 6. At the end of the discharge summary, Dr. Barbosa stated that “Discharge medication included Dilantin 75mg., b.i.d. and Tegretol 100mg., b.i.d. He is to continue on his prophylactic Factor VIII twice a week.” Id. The third document, Def.Ex. 53, is a Department of Pathology lab report entitled “LYMPHOCYTE SURFACE MARKERS (T *1491 & B CELLS)” and dated 5-13-83. Although the report gives the patient’s name as “Johnny Richards,” Dr. Barbosa testified that Def. Ex. 53 was “one of the T4 cell counts” that he had done with respect to Johnny Kellar. TTE, Barbosa CX at 21. This report gives a numerical value for “% T (E Rosettes)” as “51” in a column labeled “Peripheral Blood”; a range off to the side of the line on which this value is reported is given as “(58-82%).” Def.Ex. 53. The report then gives a numerical value for “% B (EAC Rosettes)” in the column labeled “Peripheral Blood”; a range off to the side of the line on which this value is reported is given as “(10-24%).” Id. At the bottom of the report, a handwritten “INTERPRETATION” stated the following: “T cell population is decreased slightly and B cell population is increased with increased production of immunoglos. Borderline decrease of cellular immunity.” Id. The fourth document, Def.Ex. 54, is a report prepared by Dr. Barbosa of a clinic visit by Johnny Richards on 6/22/83. It stated the following: This was a scheduled followup visit for Johnny who has Factor VIII deficiency. He underwent a T-cell subset analysis and the results are as follows: T-cell 55%; helper T-cell 33%; stvppressor T-cell 17% and 7% T-cell. Results show an elevated T suppressor cell number. This test is to be repeated in one month. Physical examination, as usual, disclosed multiple areas of ecehymosis, especially in the lower extremities. He also has some lesions such as scabies infestation for which he was given a prescription for Kwell Lotion. He was also provided with Factor VIII to be administered by his local physician, Dr. Dean Fauber, whenever the need arises. Def.Ex. 54 (emphasis added). The fifth and final document, Def.Ex. 56, is a letter from Dr. Barbosa to Dr. Dean Fauber, Johnny’s local physician, dated July 21, 1983. It stated the following: I had the pleasure of seeing Johnny today in this clinic. He is doing fairly well at this time except for his frequent episodes of soft tissue bleeding. He is receiving Factor VIII on a p.r.n. [as-needed] basis. Physical examination today disclosed, as usual, an active youngster in no acute distress. Vital signs revealed T 36.5 C, HR 104 per minute, RR 22 per minute, BP 96/66, WT 17kg., HT 95.5 cm. His ever present cervical lymphadenopathy is much better at this time. Nodes are markedly decreased in size. He was again tested for toxoplasmosis, and CMV a few weeks ago with negative results. His T-cell subset analysis showed elevated T-suppressor cell number. This rules out the possibility of AIDS. He is to be retested again in three months. Def.Ex. 56 (emphasis added). D. Discussion Viewing all the evidence, as I must, in the light and with all reasonable inferences most favorable to plaintiff, I conclude that there is substantial evidence opposed to Armour’s claim that in July 1983 Dr. Barbosa possessed independent knowledge of the AIDS risk associated with Factor VIII products sufficient to break the chain of proximate causation between Armour and Jason Christopher. Taken as a whole, Dr. Barbosa’s testimony does not as a matter of law establish that by July 1983 he had “reasonable evidence” of an “association” of an AIDS risk with Factor VIII concentrate products. When asked directly what he had heard about AIDS in 1983, Dr. Barbosa testified that it was “ill-defined,” “new to us,” and involved the fields of infectious disease, immunology, and epidemiology — fields that were not his specialties. When asked if he had modified Johnny Kellar’s Factor VIII treatment because of AIDS, Dr. Barbosa testified that he continued to use Factor VIII concentrate that was not heat-treated — i.e., that was capable of transmitting the AIDS virus — with Johnny in 1983. When asked about the point of view of hemophilia treaters early in the AIDS situation about whether what they were seeing in their patients was possibly not the same thing as was being seen in homosexuals or IV drug users but was perhaps something related to their blood products treatment, Dr. Barbosa testified that “Thinking was going in both directions” and that he did not recall what he was aware *1492 of at that time. When asked how it would have affected his treatment of Jason if he had received a warning in July 1983 from a concentrate manufacturer that there was a possible risk of AIDS transmission via Factor VIII concentrate, Dr. Barbosa testified that he would have switched Jason to cryopreeipitate, a safer treatment and one from which Jason, a moderate hemophiliac, could have benefited. I have carefully considered the evidence that Armour claims establishes Dr. Barbosa’s status as a “learned intermediary.” As previously noted, this evidence does not establish that by July 1983 Dr. Barbosa had “reasonable evidence” of an “association” of an AIDS risk with Factor VIII concentrate products. As quoted above (p. 55), Dr. Barbosa gave a single affirmative word during a spirited exchange on cross-examination to the question whether he was testing Johnny Kellar’s T-cells because he had a concern that Johnny might be in an AIDS situation as had other reported hemophiliacs. This response in and of itself does not take away the other “substantial evidence” cited above as to Dr. Barbosa’s relative lack of knowledge in July 1983 about an AIDS risk associated with Factor VIII concentrate products. At most, it introduces an inconsistency into his former testimony and creates an issue of credibility, which was properly left to evaluation by the jury. I find the written evidence concerning Dr. Barbosa’s patient Johnny Kellar of practically no persuasive force in Armour’s attempt to establish Dr. Barbosa’s alleged independent knowledge about an AIDS risk associated with Factor VIII concentrate products. The first document, the treatment record prepared by Dr. Andriola that states that Johnny is “currently no longer receiving Factor 8 due to AIDS,” at most establishes Dr. Andriola’s contemporary knowledge of an AIDS risk. Def.Ex. 5. It does not establish Dr. Barbosa’s. The second document, the 10/10/83 hospital discharge summary prepared about Johnny by Dr. Barbosa, contained no explicit mention of AIDS. Near the beginning of the document, Dr. Barbosa acknowledged that Johnny had been receiving Factor VIII concentrate on a p.r.n. [as-needed], rather than a prophylactic, basis for the past few months. In stating this fact, Dr. Barbosa conspicuously did not add, as Dr. Andriola had, that this was “due to AIDS.” Moreover, at the end of this document, Dr. Barbosa placed Johnny back on “his prophylactic Factor VIII twice a week.” Def.Ex. 6. This is hardly the course one would expect Dr. Barbosa to follow, without commenting at all about its implications with respect to an AIDS risk, if he was aware in October 1983 even that the possibility existed that AIDS could be transmitted through blood products. The third document, the Department of Pathology lab report, gave numerical values for lymphocyte surface markers. This document also contained no explicit mention of AIDS. The handwritten interpretation of Johnny’s test results merely reported that his T-eell population is decreased slightly, that his B-cell population is increased, and that there is a borderline decrease of cellular immunity. There is nothing here from which Dr. Barbosa’s alleged independent knowledge of an AIDS risk associated with Factor VIII products may be inferred. Near the beginning of the fourth document, a June 1983 report of Johnny’s clinic visit, Dr. Barbosa reported the results of a Teell subset analysis — “without any interpretation,” as he testified when asked about this document. Def.Ex. 54; TTE, Barbosa CX at 22. This document also contained no explicit mention of AIDS. Near the end of this document, Dr. Barbosa stated that Johnny was given Factor VIII to be administered by his local physician “whenever the need arises.” Def.Ex. 54. This document offers no evidence that the two topics, the T-cell subset analysis and Johnny’s Factor VIII treatment, were linked in any way at that time in Dr. Barbosa’s mind. The fifth document, a July 31, 1983, letter from Dr. Barbosa to Johnny’s local physician, contained Dr. Barbosa’s only explicit reference to AIDS in these documents. Near the end of the letter, Dr. Barbosa wrote that Johnny’s “T-cell subset analysis showed elevated T-suppressor cell number. This rules out the possibility of AIDS. He is to be retested again in three months.” Def.Ex. 56 *1493 (emphasis added). With reference to this comment, Dr. Barbosa explained during cross-examination that it “gives you an idea of the confusion at that time of the entire picture of AIDS in interpreting this test. Right now we know that the opposite is right.” TTE, Barbosa CX at 22. Near the beginning of this letter, Dr. Barbosa stated that Johnny was then “receiving Factor VIII on a p.r.n. [as-needed] basis.” Def.Ex. 56. Again, this document offers no evidence that the two topics, the T-cell subset analysis and Johnny’s Factor VIII treatment, were linked in any way at that time in Dr. Barbosa’s mind. In fact, a contrary inference could be drawn. After reporting the status of Johnny’s Factor VIII treatment, Dr. Barbosa commented upon many other factors of medical significance in Johnny’s case before mentioning the results of Johnny’s T-eell subset analysis. If being mentioned in the same document is enough to show Dr. Barbosa’s awareness of a linkage between the condition or object mentioned and Johnny’s Factor VIII concentrate, then Johnny’s vital signs, the status of his nodes, his “ever present cervical lymphadenopathy” [abnormal enlargement of the lymph glands in the neck], and the tests he was contemporaneously given for toxoplasmosis and CMV would all be candidates for such a linkage before Johnny’s T-eell subset analysis. Def.Ex. 56. In fact, that Johnny clinically presented the symptom of “ever present cervical lymphadenopathy” could explain in and of itself Dr. Barbosa’s decision to test for Johnny for AIDS, since swollen lymph glands can be a symptom of AIDS. At most, this document reflects that in mid-1983 Dr. Barbosa had some awareness of the clinical symptoms of AIDS. It clearly does not establish as a matter of law that he had the further awareness that the AIDS virus could be transmitted by Factor VIII concentrate. In short, after carefully considering all the testimony at trial, I conclude that the above evidence cited by Armour in support of its Rule 50 motion is far too ambiguous to establish that Dr. Barbosa was a “learned intermediary” who in July 1983 possessed independent knowledge of the AIDS risk associated with Factor VIII products sufficient to break the chain of proximate causation between Armour and Jason Christopher. I conclude that the above evidence cited by Armour in support of its Rule 50 motion is at best of such quality and weight that reasonable and fair-minded people in the exercise of impartial judgment (the jury in this case) might reach different conclusions. I further conclude that in July 1983 Dr. Barbosa did not, as a matter of law, possess knowledge of “reasonable evidence” of an “association” of an AIDS risk with Factor VIII concentrate. In the alternative, I conclude that in July 1983 Dr. Barbosa did not, as a matter of law, possess knowledge of the possibility that AIDS could be transmitted by blood products. At most, Dr. Barbosa’s testimony, being at times unclear and inconsistent, created an issue of credibility for the jury. See Timm v. Upjohn Co., 624 F.2d 536 (5th Cir.1980), cert. denied, 449 U.S. 1112, 101 S.Ct. 921, 66 L.Ed.2d 840 (1981), in which the court affirmed the trial court’s entry of judgment on a jury verdict finding a drug manufacturer negligent in its marketing and finding that such negligence was the proximate cause of plaintiffs injuries. The Fifth Circuit held that a rational basis for the jury’s verdict existed “since the evidence regarding the adequacy of Upjohn’s warning to Dr. Tabb was sufficient to create a question for the jury. The jury was entitled to weigh the conflicting statements made by Dr. Tabb and the other physicians along with all the other evidence presented in the case.” Id. at 539. See also Tatum v. Schering Corp., 795 F.2d 925 (11th Cir.1986), in which the court reversed the trial court’s grant of summary judgment to defendant drug manufacturer in a wrongful death action. The Eleventh Circuit held that material disputed issues of fact existed, making it incorrect to hold by summary judgment that proximate cause did not exist. Specifically, the Eleventh Circuit concluded that the testimony of the prescribing physician was “not necessarily consistent” and that “a jury could infer” from the evidence that the prescribing physician “did not possess the knowledge that he described himself as having.” Id. at 928-29. *1494 See also Therrell v. Georgia Marble Holdings Corp., 960 F.2d 1555, 1567 (11th Cir. 1992). In Therrell, the Eleventh Circuit reversed the district court’s directed verdict on an equitable accounting claim, specifically noting that “Although it appears that Dr. McCarl did lose credibility with the court, such loss of credibility is not justification for issuing a directed verdict.” Id. Determining credibility remains “a matter solely for the jury”: “ ‘[I]t is the function of the jury as the traditional finder of the facts, and not the Court, to weigh conflicting evidence and inferences, and determine the credibility of witnesses.’ ” Id. (quoting Boeing, 411 F.2d at 375). In light of this binding authority, I must deny the “learned intermediary” defense that Armour asserts on its Rule 50 motion. IV. Plaintiffs “But For” Evidence and “Cause-in-Fact” Evidence Armour claims that plaintiff failed to offer any “but for” evidence because plaintiffs counsel did not ask plaintiffs expert witnesses for their opinions as to whether Jason would have been infected with HIV in 1983 if he had received cryoprecipitate from Dr. Barbosa rather than Factor VIII concentrate. Since the risk of cryoprecipitate can depend upon where it is drawn, Armour claims that the missing testimony was critical to plaintiffs proof. Armour also claims that plaintiff failed to provide legally sufficient evidence that Armour’s Factor VIII concentrate was the “cause in fact” of Jason’s HIV-infection. Armour particularly claims that Dr. William Robinson, plaintiffs primary causation witness, did not provide substantial evidence in his testimony from which a jury could reasonably conclude that Jason had been infected with HIV from Armour’s Factor VIII concentrate. In this latter regard, Armour cites the following testimony of Dr. Robinson’s, developed on cross-examination, about the published population studies of hemophiliacs that Dr. Robinson had discussed on direct examination: 1) Dr. Robinson’s agreement that he could not predict with reasonable medical probability from such studies whether an individual patient would become HIV-infected, TT, Vol. 4 at 146-147; 2) Dr. Robinson’s agreement that there were no studies in the published medical literature discussing HIV-infection in patients who had received as little Factor VIII as Jason had, TT, Vol. 4 at 143-144, 147; 3) Dr. Robinson’s acknowledgement that Jason had received even less concentrate than all the patients in the published studies who had not been HIV-infected from Factor VIII concentrate, TT, Vol. 4 at 149; and 4) Dr. Robinson’s acknowledgement that Jason had in fact received so little Factor VIII concentrate that the probability was quite high that he would never have been infected from it at all, TT, Vol. 4 at 144, 149. Armour argues that these statements reflect that Dr. Robinson’s conclusions about the source of Jason’s HIV-infection were not based upon any scientifically reliable methodology. Armour also argues that plaintiff did not present an expert witness, as Armour did, “qualified” to assess the proper statistical use and significance of the data in Dr. Robinson’s charts. Richard Scheaffer, Ph.D., Armour’s expert witness in statistics and probability, testified as follows. First, that the only reasonable statistical inference which could be drawn from the data in Dr. Robinson’s Tables 2 and 5 (Pl.Ex. 85 and 86) was that it was exceedingly unlikely that Jason had been infected from Factor VIII concentrate because he had received so much less Factor VIII than the patients who had not been HIV-infected. Second, that it was not possible to draw statistical inferences regarding Jason’s risk of infection from cryoprecipitate from the studies reported in Dr. Robinson’s chart (Pl.Ex. 87) because there was no evidence that Jason was within a population similar to that of the study participants in relation to the cryoprecipitate he had actually received. A. Relevant Case Law In Florida, it is the plaintiffs burden to prove by a preponderance of the evidence that the defendant’s act was the cause of the injury. In other words, plaintiff must show that it is “more likely than not” that the *1495 defendant’s act was a substantial factor in bringing about the injury or that the defendant’s act probably caused the injury. A mere possibility of causation is not sufficient. See, e.g., Tampa Electric Co. v. Jones, 138 Fla. 746, 190 So. 26 (1939); Gooding v. University Hosp. Bldg., Inc., 445 So.2d 1015 (Fla.1984); Greene v. Flewelling, 366 So.2d 777 (Fla. 2d DCA 1978), cert. denied, 374 So.2d 99 (Fla.1979); Heyman v. United States, 506 F.Supp. 1145 (S.D.Fla.1981). Two Eleventh Circuit cases illustrate how a plaintiff in a case with difficult medical issues may meet her burden of proof with respect to causation through expert testimony. In Worsham v. A.H. Robins Co., 734 F.2d 676 (11th Cir.1984), plaintiff user of an intrauterine contraceptive device brought a products liability suit against the manufacturer for damages for a serious illness resulting from an infection she contracted while wearing the device and for her subsequent loss of fertility. The United States District Court for the Southern District of Florida entered final judgment reflecting a remittitur that reduced compensatory damages to $950,000 and punitive damages to $500,000. The Eleventh Circuit affirmed. In Worsham, defendant manufacturer argued that the jury could find causation only through speculation. Because the intrauterine device (IUD) she had worn had been discarded at the hospital at the time of her hysterectomy, plaintiff could not introduce it as evidence. However, plaintiff presented expert testimony that the theory of defect she argued was the cause of her infection and that the alternative causes offered by defendant manufacturer were highly unlikely. Defendant manufacturer claimed that plaintiff had failed to prove that her device was broken; that evidence introduced to support the possibility that breaks might occur in the devices showed that most of the devices had no breaks; and that plaintiff could not and did not refute possible alternative causes. Defendant manufacturer presented statistical evidence through expert witnesses that the annual incidence rate of pelvic inflammatory disease in women who do not wear IUD’s was 2-3%, and that approximately 85% of pelvic infections are in women who do not wear IUD’s. In answer to defendant manufacturer’s claims, the Eleventh Circuit stated: “What a review of the evidence reveals is that the jury was presented with a battle of experts on the question of causation.” Worsham, 734 F.2d at 682. Moreover, “As Boeing teaches us, ‘it is the function of the jury as the traditional finder of facts, and not the Court, to weigh conflicting evidence and inferences, and determine the credibility of witnesses.’ ” Id. (quoting Boeing, 411 F.2d at 375). The Eleventh Circuit then concluded that “A thorough review of the evidence presented at trial reveals that both sides presented substantial evidence on the question of causation. A jury question was, therefore, created on whether the Daikon Shield [IUD] was defective.” Worsham, 734 F.2d at 682. In Wells v. Ortho Pharmaceutical Corp., 788 F.2d 741 (11th Cir.), cert. denied, 479 U.S. 950, 107 S.Ct. 437, 93 L.Ed.2d 386 (1986), an infant and her parents brought suit against the manufacturer of a spermicide, which allegedly caused infant plaintiff to be born with birth defects. The parties waived a jury. The district court awarded $5.1 million to plaintiffs. The Eleventh Circuit affirmed with respect to defendant manufacturer’s liability, but modified the amount of damages to $4.7 million. According to the Eleventh Circuit, the Wells plaintiffs “presented well-qualified experts who testified at length concerning causation.” Id. at 744. “These experts relied on their particular areas of expertise, their personal examinations of the child ... and medical and scientific studies relative to causation to conclude that Ortho-Gynol caused Katie Wells’ birth defects. Plaintiffs presented several epidemiological studies that indicated an association between spermicide use and deleterious effects on the fetus.” Wells, 788 F.2d at 744. However, defendant manufacturer argued that plaintiffs’ expert evidence failed to focus sufficiently on epidemiology — “the field of science dealing with the relationships of the various factors which determine the frequencies and distributions of certain conditions and diseases in human populations.” Wells, *1496 788 F.2d at 744. Defendant manufacturer also challenged the district court’s credibility choices in analyzing both sides’ experts. Id. The district court had stated that it found the scientific studies presented by both sides to be inconclusive and that, in facing this “battle of the experts,” it had been forced to make credibility determinations to “decide the victor.” Wells, 788 F.2d at 745 (quoting Ferebee v. Chevron Chemical Co., 736 F.2d 1529, 1535 (D.C.Cir.), cert. denied, 469 U.S. 1062, 105 S.Ct. 545, 83 L.Ed.2d 432 (1984)). In affirming the district court’s judgment that defendant manufacturer was liable, the Eleventh Circuit recognized that a cause-effect relationship need not be clearly established by animal or epidemiological studies before a doctor can testify that, in his opinion, such a relationship exists. As long as the basic methodology employed to reach such a conclusion is sound, such as use of tissue samples, standard tests, and patient examination, products liability law does not preclude recovery until a ‘statistically significant’ number of people have been injured or until science has had the time and resources to complete sophisticated laboratory studies of the chemical. Wells, 788 F.2d at 745 (quoting Ferebee, 736 F.2d at 1535-36). The Eleventh Circuit emphasized that the district court “properly noted” that “ ‘its ultimate focus was the birth defects suffered by Katie Wells. Plaintiffs burden of proving that Katie Wells’ defects were caused by the product did not necessarily require them to produce scientific studies showing a statistically significant association between spermicides and congenital malformations in a large population.’ ” Wells, 788 F.2d at 745 (emphasis in original). The Eleventh Circuit therefore' concluded that “What matters is that this particular factfinder found sufficient evidence of causation in a legal sense in this particular case, and that that finding is not clearly erroneous.” Id. B. Evidence Plaintiff presented undisputed evidence that Jason had in his lifetime received about 100 bags of cryoprecipitate. The last eryoprecipitate that Jason received was on January 1, 1983. In July 1982 at Fort Knox, before Jason’s family moved to Florida, Jason had received three shots of Factor VIII concentrate that were not Armour’s product. All subsequent Factor VIII concentrate that Jason received was Armour’s product. After Jason’s family moved to Florida, beginning in July 1983, Dr. Barbosa treated Jason exclusively with Factor VIII concentrate. Both facilities which supplied Jason with Factor VIII concentrate in Florida, All Children’s Hospital and Bayfront Medical Center, purchased the product exclusively from Armour. (Pl.Ex. 77 gives the chronology of Jason’s receipt of various blood products.) As discussed above, Armour began marketing heat-treated (AIDS-free) Factor VIII concentrate in May 1984. Dr. Barbosa thereafter prescribed only heat-treated concentrate. Therefore, if Armour’s Factor VIII concentrate transmitted the AIDS virus to Jason, Jason must have become HIV-infected from an injection of Factor VIII concentrate that he received between July 1983 and May 1984. Based upon an examination of symptoms Jason presented clinically at the time, as documented in his medical records, two of plaintiffs expert witnesses identified an illness Jason suffered in November 1983 as a likely manifestation of primary HIV-infeetion. By November 1983, it had been over ten months since Jason had received any cryoprecipitate or any other blood product except Armour’s Factor VIII concentrate. The only blood products administered to Jason from July 1983 through November 1983 were injections of Armour’s Factor VIII concentrate. Specifically, the testimony of plaintiffs expert witnesses on these issues was as follows. Dr. Barbosa, Jason’s treating physician, testified that Factor VIII concentrate was probably the source of Jason’s HIV-infection: Q. Do you have an opinion as to the probable source of his [Jason’s] infection with the AIDS virus? A. I don’t have any reason to believe other than he got it through the infusions of factor VIII either — I just — I don’t think he got it any other way *1497 because he was not in any other high risk group. Q. As between cryoprecipitate and concentrate product, do you have an opinion as to which product would have been more likely to infect Jason with the AIDS virus? A. You know, statistically speaking, if you have a product that comes from tens of thousands of people versus a few donors, you know, I would say probably factor VIII concentrate. TT, Vol. 3 at 95-96. Dr. Barbosa then testified that the symptoms Jason presented to him clinically in November 1983 were “very suggestive” of those of primary HIV-infection. He also testified that he had never seen that “cluster of symptoms” clinically in Jason before November 1983, nor did he ever remember seeing that “cluster of symptoms” in Jason’s medical records before that time: Q. Doctor, have you read the medical literature on a condition that is referred to as primary HIV-infection? A. Yes. That is — those are the acute symptoms that a person will develop a few weeks after contracting the infection. Q. What infection? A. The HIV — after contracting the AIDS virus. Q. Did you know what that was back there in 1983? A. I don’t recall, sir. Q. What are some of the common symptoms of that condition? A. They are no different from what we see, for example in infectious mononucleosis or infectious mono, common cold, sore throat, cervical — lymph nodes enlarging in the neck, you feel achy, tired, you don’t want to do anything, runny nose. Q. When was the first time you ever saw a collection of those symptoms in Jason Christopher? A. I have to look in the records, sir. Looks like it was November 1983. Q. Were the symptoms you saw in Jason November 29, 1983, and the following day consistent with primary HIV-infection? A. From what we know now, they are very suggestive. Q. All right. Prior to that time, had you ever seen that cluster of symptoms before in Jason? A. Not in Jason, sir. Q. Have you ever seen that cluster of symptoms anywhere in his medical records prior to that time? A. Not that I can remember, sir. TT, Vol. 3 at 99-100. Plaintiff then presented both technical medical articles and expert oral testimony from William S. Robinson, M.D., 21 to establish that it was unlikely that Jason would have been infected with HIV through cryoprecipitate, had he used that blood product in 1983-84 until heat-treated (AIDS-free) Fac *1498 tor VIII concentrate became available in mid-1984. Dr. Robinson used two exhibits in particular (Pl.Ex. 87,104) to demonstrate the safety of cryoprecipitate usage in the period 1982-1985. Plaintiffs Exhibit 87 is a table prepared by Dr. Robinson in March 1991. It is entitled “Prevalence of Antibody to HIV in Hemophilia Patients Receiving Only Cryoprecipitate vs. Those Receiving Only Blood Product Concentrate.” It depicts the prevalence of HIV-infection in groups of hemophiliacs who used either cryoprecipitate or Factor VIII concentrate, as summarized from eight contrasting groupings reported by seven authors of medical studies. In half of those groups none of the cryoprecipitate users became HlV-infeeted. The highest percentage of HIV-infection in any of the groups using only cryoprecipitate was 17%. The range was 0%-17%. The percentages were 0%, 14%, 0%, 17%, 0%, 5.1%, 10%, and 0%. On the other hand, among groups of hemophiliacs who used Factor VIII concentrate, the infection rate was as high as 95%. The range was 53%-95%. The percentages were 58%, 95%, 59%, 74%, 53%, 73%, 82%, and 90%. Pl.Ex. 104 is a 1991 study by George Gjerset published in Blood, Vol. 78, No. 6, September 15), 1991, pp. 1623-1627. It is entitled “Treatment Type and Amount Influenced Human Immunodeficiency Virus Seroprevalence of Patients with Congenital Bleeding Disorders.” This study reported on a group of 282 patients with bleeding disorders, including 187 people with hemophilia A (which Jason had). Of those using only cryoprecipitate during the early 1980’s, before Factor VIII concentrate was heat-treated (AIDS-free), only 14% became HIV-infected. On the other hand, 88% of those using Factor VIII concentrate during that period became HIV-infected. Those patients in the study who used low amounts of cryoprecipitate were not HIV-infected. These patients who were not HIV-infected used the same low levels of cryoprecipitate that Jason would have used had his physician prescribed cryoprecipitate rather than Factor VIII concentrate for him. The studies summarized or reported in Pl.Ex. 87 and 104 formed part of the basis for Dr. Robinson’s opinion that use of cryoprecipitate carried a low risk of HIV-infection in the early 1980’s. In discussing Pl.Ex. 104, Dr. Robinson testified as to the “particular reason” that the Gjerset study was “helpful in trying to determine the source and time” of Jason’s HIV-infection: “Well, I think it is important because what it does is tell us the probability of infection of patients that use[d] cryoprecipitate only in terms of the quantity of cryoprecipitate they used.” TT, Vol. 4 at 74. Dr. Robinson then explained why this study suggested that the probability that Jason was infected with cryoprecipitate was “low”: Q. All right. Where does the first HIV-positive person surface on this particular diagram? A. Well, among the hemophilia patients, no patient that got less than 500 bags became antibody-positive [HIV-positive], was infected with HIV. Among the von Willebrand’s patients, no patient that received less than 300 bags of cryoprecipitate became antibody-positive. Q. Did any of the hemophiliacs or the people with von Willebrand’s disease who received cryo[precipitate] in the same amount as Jason received in his lifetime become infected with the AIDS virus? A. None did in either group. Q. All right. Do you feel that the information that’s in this graph is helpful in trying to determine, number one, the source of Jason’s HIV-infection; and, number two, the most probable time of his infection. A. Well, it suggests that with — the probability that he was infected with cryoprecipitate is low. It’s — and if he was infected with cryoprecipitate, it would have been, of course, during the time he took it in 1982; he took most of it. But the chance or probability that he was infected by that would appear from data like this to be low. TT, Vol. 4 at 76. Dr. Robinson also testified that the symptoms that Jason presented clinically to Dr. Barbosa on November 29-30, 1983, as docu *1499 mented in Jason’s medical records, appeared “at a time that’s consistent with the incubation period of primary HIV-infection” (TT, Vol. 4 at 84-85): A. ... I think the other thing that would be consistent is that he did receive Factor VIII concentrate prior to that illness at a time that’s consistent with the incubation period of primary disease with HIV-infection. Q. All right. In that regard, let me show you Plaintiffs Exhibit Number 77, which is in evidence. This is the list of Jason’s blood product usage with the dates, and looking at this list, you just indicated that he had received some concentrate product before this illness in late November 1983. Which particular Factor VIII concentrate injections are you talking about? A. Well, if you look at many descriptions of the primary disease in the literature, it usually occurs in a window between, say, one week, and six weeks after exposure to the virus or infection with the virus. And Jason received a Factor VIII concentrate 19 days before this illness described in the medical records. That would be consistent with the time interval between infection and the primary illness. TT, Vol. 4 at 86-87. Dr. Robinson further testified that he had reviewed Jason’s clinical records of an illness Jason had suffered in July 1982, shortly after receiving three shots of Factor VIII concentrate. This was the only occasion upon which Jason received Factor VIII concentrate that was not Armour’s product. After receiving this Factor VIII concentrate, Jason’s medical records document that he presented clinical symptoms of a mild fever, a yellowing of the whites of his eyes, and an enlarged liver. When asked if these symptoms were consistent with a particular type of viral infection, Dr. Robinson stated that the symptoms were consistent with an infection that could be caused by a number of viruses, but that such symptoms were most often caused by hepatitis viruses. TT, Vol. 4 at 82. When asked whether these symptoms would represent primary HIV-infection in Jason at that time, Dr. Robinson stated “No. I think there’s no evidence that this represents primary HIV-infection. But that’s based not only on the symptoms, it’s based on the laboratory findings at the time and other data besides that.” TT, Vol. 4 at 82. The reason that Dr. Robinson was certain that this illness “undoubtedly” represented an acute case of hepatitis B which Jason had “probably” contracted from his Factor VIII concentrate was that lab reports in August 1982 gave “absolute” evidence that Jason was at that time infected with the hepatitis B virus. TT, Vol. 4 at 82-84. Dr. Robinson stated that the measurements of Jason’s liver injury reflected in these lab reports showed “very, very high” values “never” seen in HIV-infection. TT, Vol. 4 at 83-84. In the following testimony, Dr. Robinson summarized various medical reasons supporting his conclusions that the source of Jason’s HIV-infection was the Factor VIII concentrate he received and that November 1983 was the likely time of Jason’s HIV-infection: Q. Based upon all the evidence we’ve reviewed, your experience with HIV and AIDS, and the medical literature which you have familiarized yourself with over the years, do you have an opinion within a reasonable degree of medical probability as to the probable source of Jason’s HIV-infection? A. Well, I would think the most probable source would be the Factor VIII concentrate that he received because that carries a higher risk for infection that cryoprecipitate from individual donors. Q. All right. Doctor, for the purposes of the next question, I want you to assume that Jason was switched to heat-treated Factor VIII concentrate by Dr. Barbosa sometime shortly after May 15, 1984; is that understood? A. Yes. Q. Based upon the medical evidence we’ve reviewed today, your reading of the literature and your own experience with HIV and AIDS, do you have an opinion within a reasonable degree of medical probability as to the probable time period during which Jason Christo *1500 pher was initially infected with HIV, the AIDS virus? A. Well, he received such small quantities that I would think that the most probable time is the period immediately before switching to safe Factor VIII concentrate, heated Factor VIII concentrate, and that would be in late ’83 or early ’84. And the probability that he was infected earlier I would think would be significantly lower. TT, Vol. 4 at 88-89. Q. Now, I want you to look at the blood product usage chart there that’s in evidence, and I’d like to ask you do you have an opinion within a reasonable degree of medical probability as to which injections of Factor VIII concentrate that are shown on this chronology in your opinion would be the injection or injections which most likely would have infected Jason Christopher? A. Well, I don’t think there’s any way to identify a lot of Factor VIII or a time when this child was treated with Factor VIII that can prove that that was the Factor VIII that infected the child. There’s no way that that could be done, but I think that, as I mentioned this morning, there were ways of determining a probable time of infection, and I think one of those relates to the quantity of Factor VIII that he received. He [Jason] received a very small dose [of Factor VIII concentrate] — that would suggest he was probably infected late in the course of this therapy in terms of years before he switched to heat-treated. And, as I said this morning, that would be late ’83 or 1984, I would guess. I think if you look at the amounts of Factor VIII concentrate that he received, he received none in ’81; can’t implicate ’81. And then each year ’82, ’83, ’84, he received increasing amounts — less than a thousand units in ’82; 1.9 or 1,900 in ’83; and even more in ’84, 2,640. That would suggest that the risk of infection from Factor VIII concentrate would be increasing with respect to the amount of Factor VIII that he got. You [sic] would be more likely to be infected in ’84, at least before he switched to heat-treated because he was getting more Factor VIII concentrate that year than in ’83 and more in ’83 than ’82. So that’s another factor that goes into my strong feeling that the probability is that he was infected in late ’83 or ’84. In addition to that, the only illness that I would find in the medical records that’s consistent with primary infection was November 1983, the end of 1983. That illness is perfectly consistent with primary infection. We can’t prove that it was, but it was consistent. And if you look at the Factor that he received, the Factor VIII material that he received before that illness in November of ’83, that’s consistent with the incubation period of primary disease. It would be the Factor VIII concentrate that he got on November the 10th; 19 days later he had this illness. That’s consistent with the primary illness. He got no cryoprecipitate within 11 months of that. So if that was the primary illness related to HIV-infection, you can’t implicate cryoprecipitate in that illness. But there was Factor VIII concentrate therapy that was given three weeks or 19 days before that illness, and that’s perfectly consistent with the incubation period of infection to primary illness, and that then is in the time frame, end of 1983, that I conclude is the most probable time of infection from the quantities of Factor VIII concentrate that he received each year, and the fact that he got increasing amounts through ’82, ’83 and ’84. TT, Vol. 4 at 97-99. Dr. Robinson thus specified November 1983 as the “most probable” time of Jason’s HIV-infection based upon his review of Jason’s blood products usage; the clinical symptoms that Jason presented during various illnesses, as documented in his medical records; and the known incubation period of primary HIV-infection. *1501 C. Discussion Armour’s claim that plaintiff failed to offer any “but for” evidence misconceives the nature of plaintiffs burden in this case. Armour argues that plaintiff failed to offer such evidence because plaintiffs counsel did not ask plaintiffs expert witnesses for their opinions as to whether Jason would have been infected with HIV in 1983 if he had received cryoprecipitate from Dr. Barbosa rather than Factor VIII concentrate. However, it was not plaintiffs burden to prove that Jason could not possibly have been infected with the AIDS virus had he used only cryoprecipitate in 1983. It was rather plaintiffs burden to prove that it was more likely than not that, but for Armour’s failure to provide a timely or an effective warning about the AIDS risk associated with its Factor VIII concentrate, Jason would not have become HIV-infected. Plaintiff presented substantial evidence, some of which is quoted above, to satisfy this burden. Dr. Barbosa, Jason’s treating physician, testified that, had he received a warning from a concentrate manufacturer in July 1983 that there was a possible risk of AIDS transmission via Factor VIII concentrate, he would have changed Jason’s therapy by switching him to cryoprecipitate. Dr. Barbosa explained that, since Jason was a moderate hemophiliac, he could have benefited from using cryoprecipitate. In addition, he emphasized that cryoprecipitate “comes from one single person and those are volunteer donors.” Dr. Barbosa therefore would have felt that cryoprecipitate was “a safer way to treat” Jason had he received a warning from a concentrate manufacturer in July 1983 that there was a possible risk of AIDS transmission via Factor VIII concentrate. TT, Vol. 3 at 90-92. In addition to the testimony of Dr. Barbosa, plaintiff presented considerable other evidence, in the form of both oral testimony and of scientific articles published in medical journals, to establish that in 1982-84 there was a significantly smaller probability that treatment with cryoprecipitate — as opposed to treatment with Factor VIII concentrate— would transmit the AIDS virus. Dr. Robinson, plaintiffs expert witness who has for over twenty-five years served on the faculty of the Stanford University Medical School as a professor of infectious diseases, testified that the probability of Jason’s having been infected with cryoprecipitate was “low.” TT, Vol. 4 at 76. Dr. Robinson particularly used the epidemiological studies referenced in Pl.Ex. 87 and 104 to discuss the scientific consensus that exists to support such a conclusion. PI. Ex. 87 is a table prepared by Dr. Robinson. It depicts the prevalence of HIV-infection in groups of hemophiliacs who used either cryoprecipitate or Factor VIII concentrate, as summarized from eight contrasting groupings reported by seven authors of medical studies. In these studies, the range of HIV-infection among cryoprecipitate users was 0-17%. On the other hand, the range of HIV-infection among Factor VIII concentrate users was 53-95%. Similar results were obtained in Pl.Ex. 104, the 1991 Gjerset study published in Blood. Gjerset reported that only 14% of cryoprecipitate users became HIV-infected. On the other hand, 88% of Factor VIII concentrate users became HIV-infected. In addition, those cryoprecipitate users who were not HIV-infected used the same low levels of cryoprecipitate that Jason would have used had his physician prescribed cryoprecipitate rather than Factor VIII concentrate for him. I conclude that presenting the above evidence satisfied plaintiffs burden to present sufficient evidence of “but for” causation to create a jury question. Armour also claims that plaintiff failed to provide legally sufficient evidence that Armour’s Factor VIII concentrate was the “cause in fact” of Jason’s HIV-infection. Armour particularly claims that Dr. Robinson’s testimony did not provide substantial evidence from which a jury could reasonably conclude that Jason had been infected with HIV from Armour’s Factor VIII concentrate. With respect to causation in fact, it was plaintiffs burden to prove that it was more likely than not that Jason was infected with the AIDS virus through using Armour’s Factor VIII concentrate. Plaintiff presented substantial evidence, some of which is quoted *1502 above, to satisfy this burden. Jason suffered an illness in November 1983 that both Dr. Barbosa, Jason’s treating physician, and Dr. Robinson, plaintiffs primary causation witness, identified as a likely manifestation of primary HIV-infection. Indeed, Dr. Robinson testified that this was the “only” illness in Jason’s entire medical records “consistent with” primary HIV-infection. TT, Vol. 4 at 98. Dr. Robinson had earlier testified that, based on the results of contemporary lab reports, the illness that Jason suffered in July 1982 while at Fort Knox, was “undoubtedly” an acute ease of hepatitis B. TT, Vol. 4 at 84. According to Dr. Robinson, the known incubation period of primary HIV-infection is between one and six weeks after infection with the virus. Armed with this information, Dr. Robinson testified that the “most probable time” of Jason’s HIV-infection was November 10, 1983. TT, Vol. 4 at 98-99. On that date, Jason had received Factor VIII concentrate. Nineteen days later, Jason presented symptoms of primary HIV-infection. It was undisputed that the Factor VIII concentrate that Jason received on November 10, 1983, was Armour’s product. It is undisputed that Jason had received no other blood products of any kind, except for Armour’s Factor VIII concentrate, for over ten months before November 10, 1983. In summary, when Dr. Robinson was asked if he had an opinion “within a reasonable degree of medical probability” which injection or injections of Factor VIII concentrate most likely infected Jason with the AIDS virus, Dr. Robinson identified the injection given on November 10, 1993. TT, Vol. 4 at 98-99. Further, in his answer, Dr. Robinson unquestionably identified and discussed the particular combination of factors that, when considered together, led to his expert opinion that November 1983 was “the most probable time of infection.” This particular combination of factors included 1) the quantities of Factor VIII concentrate that Jason received in various years; 2) the chronology or pattern of Jason’s increasing use of Factor VIII; 3) the fact that the “only” illness in Jason’s medical records “consistent with” primary HIV-infection occurred in November 1983; and 4) the fact that this illness occurred 19 days after Jason received Factor VIII concentrate, which was “perfectly consistent with the incubation period of infection to primary illness.” Id. In considering Dr. Robinson’s testimony about this particular combination of factors, taken as a whole, the jury would have been justified in concluding that it was Dr. Robinson’s expert opinion, within a reasonable degree of medical probability, that it was the Factor VIII concentrate administered on November 10, 1983, that infected Jason with HIV. Further, I reject Armour’s argument that Dr. Robinson’s conclusions about the source of Jason’s HIV-infection were not based upon any scientifically reliable methodology. First, as is evident from the following testimony bearing on the appropriate use of statistical data presented in medical epidemiological studies, Dr. Robinson vehemently disagreed with Armour’s characterization of Jason’s “risk” of HIV-infection as 100%: Q. But we know that Jason Christopher was infected so, in fact, his chance of infection from that perspective was a 100 percent? A. It was not. Q. And you cannot go back and look at the data regarding study population and say that some particular patient in there is or is not more probably than not going to become infected or at any given time; isn’t that true? A. I think you’re wrong. I think the risk of his infection, ongoing all the time until he was infected, was determined by the amount of Factor VIII he got. It was small amounts, so his risk was very low during ’82, ’83, ’84. He had a small risk. And just because he happened to be one of the few who was infected doesn’t mean that the risk was high. It just does not mean that. You don’t understand what risk means. Q. I suppose not. A. You don’t. Q. If you took one of those samples that you got up there — let’s take the Ragni 10 out of 59 cryoprecipitate users, okay. And if you got the 59 people standing in front you [sic] and one of them is named *1503 Jason Christopher, you don’t have any idea whether if [sic] he’s going to be one of those 59 — excuse me, one of the 17— excuse me, one of the 10 that got infected, or one of the balance of the population who didn’t get infected; isn’t that right? A. But I can estimate the risk based upon how much .he’s getting. Q. But— A. If he’s getting a small amount, then the risk is low. If he’s getting a large amount, then the risk is higher. And that risk remains low or high. And whether or not he subsequently does become infected does not change that risk. If he’s one of the unfortunate few to be infected, even that doesn’t change his risk of infection before he is infected if he’s in a group that receives small amounts of Factor VIII. It doesn’t affect the risk. TT, Vol. 4 at 144-146. As developed by Armour on cross-examination, Dr. Robinson’s explanation of the reason for his vehement disagreement with Armour’s characterization of Jason’s “risk” of HIV-infection as 100% illustrates some of the statistical anomalies presented by this case. So does Armour’s caviling about certain other of Dr. Robinson’s statements elicited on cross-examination, such as Dr. Robinson’s agreeing that there were no studies in the published medical literature discussing HIV-infection in patients who had received as little Factor VIII concentrate as Jason had, TT, Vol. 4 at 143-144. According to expert evidence admitted at trial, Jason received less eryoprecipitate than other patients who had not been HIV-infected. Jason also received less Factor VIII concentrate than other patients who had not been HIV-infected. Yet it was the tragedy of this child’s situation and it is the heart of this case that Jason was in fact HIV-infected even though he apparently had only a low “risk” of being HIV-infected. As the Eleventh Circuit stated in Wells v. Ortho Pharmaceutical Corp., 788 F.2d 741 (11th Cir.), cert. denied, 479 U.S. 950, 107 S.Ct. 437, 93 L.Ed.2d 386 (1986), a “distinction exists between legal sufficiency and scientific certainty.” Id. at 745. In such a case, the Eleventh Circuit emphasized that the fact finder’s “ultimate focus” must be on the injuries suffered by the plaintiff. Moreover, a plaintiffs burden of proving that her injury was caused by the product does “not necessarily require” her “to produce scientific studies showing a statistically significant association” between the product and the injury “in a large population.” Id. In Wells, in fact, the district court stated that it, as fact finder, had found the expert studies to be inconclusive on the ultimate issue of whether the product caused plaintiffs injuries. It was therefore forced to make credibility determinations to “decide the victor” in this “battle of the experts.” Id. (quoting Ferebee, 736 F.2d at 1535). Nevertheless, the Eleventh Circuit affirmed the conclusion of the district court that defendant manufacturer was liable for plaintiffs injuries. Plaintiff here presented Dr. Robinson, a well-qualified expert, a Stanford University Medical School professor of infectious diseases, to interpret the significance of the results reported in epidemiological studies published in professional medical journals with respect to plaintiffs burden of proof in this case. On the other hand, Armour presented Dr. Richard Scheaffer, a well-qualified expert, a University of Florida professor of statistics, 22 to interpret the reasonable statistical inferences that should be drawn from *1504 these studies. Dr. Robinson, the medical expert, had no formal training in statistics. Dr. Schaeffer, the statistics expert, had no formal training in medicine. 23 I conclude, as did the Eleventh Circuit in Worsham v. A.H. Robins Co., 734 F.2d 676 (11th Cir.1984), that “what a review of this evidence reveals” is “that the jury was presented with a battle of experts on the question of causation.” Worsham, 734 F.2d at 682. Moreover, “[a]s Boeing teaches us, ‘it is the function of the jury as the traditional finder of facts, and not the Court, to weigh conflicting evidence and inferences, and determine the credibility of witnesses.’ ” Id. (quoting Boeing, 411 F.2d at 375). Second, and more importantly, I note that plaintiffs “cause-in-fact” evidence was not confined to expert testimony interpreting epidemiological studies, but was also focused, in the testimony of both Dr. Barbosa and Dr. Robinson, on the child’s own history of blood products usage and clinical symptoms as documented in his medical records. In the case of Dr. Barbosa, Jason’s treating physician, plaintiffs “cause-in-faet” evidence included his testimony about clinical symptoms presented to him by Jason on November 29, 1983, as Dr. Barbosa contemporaneously documented these symptoms in Jason’s medical records. Dr. Barbosa further testified that, “from what we know now,” these symptoms were “very suggestive” of primary HIV-infeetion. TT, Vol. 3 at 100. In the case of Dr. Robinson, plaintiffs primary causation witness, plaintiffs “cause-in-fact” evidence included his testimony that this November 1983 illness was the “only” illness that he found in Jason’s medical records “that’s consistent with primary [HIV-]infection.” TT, Vol. 4 at 98. As discussed above, Dr. Robinson specified November 1983 as the “most probable” time of Jason’s HIV-infeetion based upon Dr. Robinson’s review of Jason’s blood products usage; the clinical symptoms that Jason presented during various illnesses, as documented in his medical records; and the known incubation period of primary HIV-infeetion. TT, Vol. 4 at 97-99. In presenting this evidence, I note that plaintiff here has presented the same kind of “cause-in-fact” evidence that the Eleventh Circuit in Wells found sufficient to affirm the fact finder’s judgment that defendant manufacturer was liable for plaintiffs injuries. Plaintiff “presented well-qualified experts who testified at length concerning causation.” Wells, 788 F.2d at 744. These experts “relied on their particular areas of expertise, their personal examination of the child [at least by Dr. Barbosa], and medical and scientific studies relative to causation to conclude” that the product caused plaintiffs injuries. Id. In addition, plaintiff “presented several epidemiological studies that indicated an association between” use of the product and the injury. Id. In light of the binding authority cited above, I conclude, as the Eleventh Circuit did in Worsham, that “A thorough review of the evidence presented at trial reveals that both sides presented substantial evidence on the question of causation.” Worsham, 734 F.2d at 682. A jury question was therefore created on whether plaintiff presented sufficient “but for” evidence and “cause-in-fact” evidence to prove that it was more likely than not that Armour’s Factor VIII concentrate infected Jason with the AIDS virus. Accord Boeing Co. v. Shipman, 411 F.2d 365 (5th Cir.1969) (en banc); Worsham v. AH. Robins Co., 734 F.2d 676 (11th Cir.1984); Wells v. Ortho Pharmaceutical Corp., 788 F.2d 741 (11th Cir.1986). V. Conclusion I therefore deny Armour’s renewed Rule 50 motion for judgment as a matter of law. Footnotes [1] See also the following Florida decisions in which the courts held that "failure-to-warn” claims were subject to Florida’s products liability statute of limitation: Eddings v. Volkswagenwerk, *1471 A.G., 635 F.Supp. 45 (N.D.Fla.1986), aff'd, 835 F.2d 1369 (11th Cir.), cert. denied sub nom. Griffin v. Ford Motor Co., 488 U.S. 822, 109 S.Ct. 68, 102 L.Ed.2d 44 (1988); Wallis v. Grumman Corp., 515 So.2d 1276 (Fla.1987); Williams v. American Laundry Machinery Industries, 509 So.2d 1363 (Fla. 2d DCA 1987), review denied, 525 So.2d 881 (Fla.1988). See also Babush v. American Home Products Corp., 589 So.2d 1379 (Fla. 4th DCA 1991), review denied, 599 So.2d 1278 (Fla.1992); University of Miami v. Bogorff, 583 So.2d 1000 (Fla.1991). [2] Under Florida law, the running of the statute of limitation in a personal injury action is not tolled during a plaintiff's minority. Fla.Stat. § 95.051. [3] See also Durden v. American Hosp. Supply Corp., 375 So.2d 1096, 1099 (Fla. 3d DCA 1979) (per curiam) (approved in Silva), cert. denied, 386 So.2d 633 (Fla.1980), in which the court refused to apply the medical malpractice statute of limitations to an action by a donor against a blood bank. See also Rostocki v. Southwest Florida Blood Bank, Inc., 276 So.2d 475 (Fla.1973), in which the court held that, where the action arose before passage of § 672.316(5), the sale of blood constituted sale of a product, and defendant blood bank was subject to strict liability. See also Williamson v. Memorial Hospital of Bay County, 307 So.2d 199 (Fla. 1st DCA 1975), in which the court held that, under § 672.316(5), a plaintiff could maintain an action on the theory of implied warranty against the hospital and physicians who sold and administered infected blood to her only if she alleged and proved that the defect was detectable or removable by the use of reasonable scientific procedures or techniques. See also Doe v. American Nat'l Red Cross, 798 F.Supp. 301 (E.D.N.C.1992) (applying North Carolina law) and Kaiser v. Memorial Blood Center, Inc., 486 N.W.2d 762 (Minn. 1992) (applying Minnesota law), in which the courts did not rule on the purpose of Fla.Stat. § 672.316(5), since that issue was not before them. [4] In this discussion, the Sicuranza court reviewed in some detail its earlier decision in Williamson v. Memorial Hospital of Bay County, 307 So.2d 199 (Fla. 1st DCA 1975). In Sicuranza, the court, while clarifying a possible ambiguity in its earlier decision, reiterated that Williamson, "when read in its entirety, properly construes section 672.316(5) as pertaining only to allegations of breach of implied warranties of fitness and merchantability.” Sicuranza, 582 So.2d at 56 (emphasis added). [5] In arguing that this action is time-barred under the negligence statute of limitation, Armour relies upon Florida case law stating that a negligence cause of action accrues when a plaintiff has notice of either the negligence or the injury (citing Johnson v. Deluxe Tire Service, Inc., 544 So.2d 1158, 1160 (Fla. 5th DCA 1989) (automobile accident case); Almengor v. Dade County, 359 So.2d 892, 894 (Fla. 3d DCA 1978) (medical malpractice case); University of Miami v. Bogorff, 583 So.2d 1000, 1002 (Fla.1991) (medical malpractice part of a case that also included a products liability claim); Barron v. Shapiro, 565 So.2d 1319 (Fla.1990) (medical malpractice case)). This rule did not require that a plaintiff know of a possibly causal connection between the injury and the negligence in order to trigger running of the limitations period. This rule was recently superseded in medical malpractice cases in Tanner v. Hartog, 618 So.2d 177 (Fla.1993), as described below. The rule now requires that a plaintiff must know of a possibly causal connection between the injury and the negligence in order to trigger running of the limitations period. In answer to a question certified to it by Florida’s Second District Court of Appeal, the Supreme Court of Florida in Tanner held that the stillbirth of a child was not such an obvious injury as to place a plaintiff on notice of the possible invasion of her legal rights to begin running of the statute of limitations. Tanner, 618 So.2d at 182. In Tanner, the Supreme Court of Florida changed its former rule, holding "that the knowledge of the injury as referred to in the rule as triggering the statute of limitations means not only knowledge of the injury but also knowledge that there is a reasonable possibility that the injury was caused by medical malpractice." Tanner, 618 So.2d at 181 (emphasis added). In no way can this case give any support to Armour’s claim. Quite the contrary. [6] The Bogorff plaintiff brought both medical malpractice and products liability claims. [7] I note that the Supreme Court of Florida’s latest ruling on such issues in Tanner now mandates that plaintiffs in medical malpractice actions must have a similar level of knowledge of possible causation that the Babush court had previously stated plaintiffs in products liability actions must have before a court should determine that the statute of limitations begins to run — i.e., plaintiffs must know that the connection between the injury and the negligence is possibly causal. Tanner, 618 So.2d at 181-82; Babush, 589 So.2d at 1381. [8] He became "paralyzed and brain-damaged." Bogorff at 1004. [9] The testimony of both parents as to when they might have first learned of an association between AIDS and hemophilia was not always precise when asked about dates in conjunction with the dates of moves, divorces, or marriages. For instance, when Jason’s father was asked if he first saw a television show featuring people with hemophilia and AIDS shortly after his remarriage in early 1985, he said "Yes.” Trial Transcript ("TT”), Vol. 3 at 28-29. However, when asked if he had ever discussed with Dr. Barbosa, Jason's treating physician, the possible risks of transmitting AIDS through blood products, he said, "That would have been shortly after the TV program, yes” and "that had to been [sic] about ’88; Jason was already infected.” TT, Vol. 3 at 33. Similarly, when Jason’s mother was asked when she first became aware of the possibility that AIDS could be transmitted by blood or blood products, she said, "It was after we came to Florida. I don’t remember when.” When then asked how long after she came to Florida, she said, "A year maybe.” TT, Vol. 5 at 114. However, when she was asked if, before being informed that Jason had tested positive, she had discussed a relationship between Factor VIII concentrate and AIDS with Dr. Barbosa, she said, "No, sir, ’cause before Jason was living with me, his dad took him to the doctor 99 percent of the time. I was — you know, I wasn't doing that. I must — I don’t know — maybe I was thinking of then.” TT, Vol. 5 at 119. Despite such imprecision as to some dates, I concluded that both parents were clear and credible in their testimony that they first learned of their son’s HIV-infection in 1988. I reached my conclusion that Jason’s parents had no actual knowledge sufficient to begin running of the statute of limitations in this action before December 27, 1985, after carefully considering all of the testimony at trial. In any event, the cumulated testimony of the parents created a legitimate factual jury question, which was decided unanimously against Armour's position. [10] ' See also Norsworthy v. Holmes Regional Medical Ctr., Inc., 598 So.2d 105 (Fla. 5th DCA 1992), approved sub nom. Kronman v. Norsworthy, 618 So.2d 186 (Fla.1993), in which a child hospitalized for difficulty breathing was left with subglottic stenosis (the narrowing of the airway below the vocal cords) after treatment that included intubations and a tracheotomy. In Norsworthy, the appellate court reversed the circuit court, holding that a jury question was presented as to whether the parents were on notice of the incident’s giving rise to medical malpractice at the time they first learned of their son’s condition: "There is little, if anything, in this record to suggest that the ‘injury’ was the result of anything other than natural consequences of a recognized medical treatment competently performed.” Norsworthy, 598 So.2d at 108. See also Steiner v. Ciba-Geigy Corp., 364 So.2d 47 (Fla. 3d DCA 1978), cert. denied, 373 So.2d *1478 461 (Fla.1979), in which plaintiff brought an action against a drug manufacturer for his loss of eyesight that occurred after taking the manufacturer’s drug. In Steiner, the appellate court affirmed the circuit court’s ruling that the action was time-barred because the action was filed more than four years after the trauma of plaintiff's loss of eyesight and the surrounding circumstances of his taking the drug should have made plaintiff realize that the drug manufacturer might be legally responsible. See also Frankowitz v. Propst, 489 So.2d 51 (Fla. 4th DCA), review denied, 492 So.2d 1334 (Fla.1986), in which a plaintiff who brought a medical malpractice action against certain doctors later amended her complaint, after running of the two-year statute of limitations, to add another doctor as defendant. In Frankowitz, the appellate court reversed the circuit court, holding the action time-barred as to the added defendant because the means of discovering this doctor’s involvement were readily available through the medical records and the amended complaint did not relate back to the original complaint that was timely filed. See also MacMurray v. Board of Regents, 362 So.2d 969 (Fla. 1st DCA 1978), cert. denied, 370 So.2d 460 (1979), in which plaintiff did not file suit until May 1976. However, in late 1973, plaintiff knew that she had Hodgkin's disease, and plaintiff's, husband, a physician, had acquired plaintiff's X-rays and other medical records from the hospital. These records contained evidence of an error in interpreting plaintiff's X-rays that defendant admitted constituted actionable negligence. In MacMurray, the appellate court affirmed the circuit court’s ruling that the action was time-barred because plaintiff had the medical records under her control whereby an earlier and timely examination would have made available the information needed to file her action before running of the medical malpractice statute of limitations. [11] After Tanner, the rule in medical malpractice cases is that, if "the nature of the injury, standing alone," communicates "the possibility of medical negligence,” then the statute of limitations "will immediately begin to run upon discovery of the injury itself.” Tanner, 618 So.2d at 181-82. On the other hand, "if the injury is such that it is likely to have occurred from natural causes,” the statute will not begin to run until "there is reason to believe that medical malpractice may possibly have occurred." Id. (emphasis added). [12] In noting that there was no evidence in this case that the records in question were available to Jason’s parents, I stated in my bench ruling that I found the fiction that such records were available to the patient or the patient’s family hard to accept. With reference to this statement, Armour argues in its brief that the court must not permit its personal views to influence the outcome of this diversity action. While I may have been stating my personal views, I of course did not permit them to influence the outcome of this diversity action. At the time of my bench ruling, I was beginning to make a distinction about what the Nardone court meant, in the context of the facts of that case, by its statement that knowledge of the medical records “of a character as to be obtainable by, or available to, the patient" should be imputed to a minor plaintiff's parents. Id. This distinction is more fully elucidated above. [13] The other is "exposure to the product in question." Id. [14] The second such factor is “knowledge that the particular drug had been administered.” Id. Because both of Jason’s parents knew that their son was receiving Factor VIII concentrate, I conclude that this factor is met by the facts of this case. The third such factor, discussed below, is "constructive knowledge of medical opinion in the hospital records that the drug may have contributed to the injury.” Id. [15] When I stated in my bench ruling that Def. Ex. 52B was a lab report with the percentage, 17 percent, that was an indication of a diagnosis of AIDS and that Def. Ex. 52C was a lab report showing a positive test for AIDS, my characterizations of the information contained in these documents reflected my own present understanding of their contents. It did not reflect the understanding that I expected lay people who read these documents in May 1985 to have had at that time. [16] Armour may have been misled in its understanding of the essential nature of plaintiffs claim under the comprehensive framework of federal law applicable to this case by its exclusive reliance on failure-to-warn cases in which an allegedly inaccurate warning was in fact given. However, as is apparent by an examination of the cases listed below, even the cases cited by Armour in support of its position provide little, if any, support to Armour’s position. See Felix, 540 So.2d 102, 104 (Fla.1989), discussed above, in which the Supreme Court of Florida approved the lower court's grant of summary judgment to defendant drug manufacturer, who "could not be penalized for the failure of the doctor to impart knowledge concerning the dangers of the drug of which the doctor had been warned and was aware" (emphasis added). See Zanzuri, 748 F.Supp. 1511 (S.D.Fla.1990), discussed above, in which the court denied defendant drug manufacturer summary judgment where the prescribing physician stated that he relied heavily upon the drug manufacturer’s literature and where the warning issued by the manufacturer contained misstatements. See also Tatum v. Schering Corp., 795 F.2d 925, 928-29 (11th Cir.1986), in which the Eleventh Circuit reversed the district court's grant of summary judgment to defendant drug manufacturer. In Tatum, the Eleventh Circuit concluded that the testimony of the physician who received an allegedly inadequate warning about a drug he prescribed was “not necessarily consistent.” A jury could therefore "infer” that the physician "did not possess the knowledge he described himself as having.” See also Windham v. Wyeth Laboratories, Inc., 786 F.Supp. 607, 612 (S.D.Miss.1992), in which the court granted defendant drug manufacturer summary judgment. In Windham, the court stated that the allegedly inadequate warnings could not have been the proximate cause of plaintiff's injuries where the prescribing physician testified that he considered the risks associated with using the drug before prescribing it and that, even if provided with a fuller warning issued at a later date, he would have still prescribed the drug. However, the court noted that it would have "reservefd] the issue of credibility for the jury’s determination” without this "unequivocal” testimony that the physician would have prescribed *1486 the drug even if he had received the later warning. Armour moreover cites in support of its position the following irrelevant cases not involving prescription drugs or biological products regulated by the comprehensive framework of federal law applicable to the case presently before me: Wickham v. Baltimore Copper Paint Co., 327 So.2d 826 (Fla. 3d DCA 1976), cert. denied, 339 So.2d 1173 (Fla.1976); Willett v. Baxter Int'l, Inc., 929 F.2d 1094, 1099 (5th Cir.1991); Kirsch v. Picker Int’l, Inc., 753 F.2d 670, 672 (8th Cir. 1985); Lopez v. Southern Coatings, Inc., 580 So.2d 864 (Fla. 3d DCA 1991); Ashby Div. of Consol. Aluminum Corp. v. Dobkin, 458 So.2d 335 (Fla. 3d DCA 1984). [17] In the following cases involving inadequate warnings, the courts also addressed the issue of when defendant manufacturer's duty to warn had arisen. See also Hurley v. Lederle Laboratories Div. of American Cyanamid Co., 863 F.2d 1173 (5th Cir.1988), in which the Fifth Circuit reversed the district court's grant of summary judgment to defendant vaccine manufacturer. In Hurley, parents of an infant who suffered an adverse reaction to a vaccine alleged that the prescribing physician received an inadequate warning about its dangers. The court denied defendant manufacturer summary judgment because it concluded that a factual issue remained "as to whether the manufacturer provided the FDA all the necessary and available information on which to base the warning.” Id. at 1179. See also Feldman v. Lederle Laboratories, 125 N.J. 117, 592 A.2d 1176 (1991), in which, in a procedurally complex case, the Supreme Court of New Jersey reversed and remanded the appellate court’s reversal of the trial court's entry of judgment in favor of a patient who suffered tooth discoloration as a result of taking a tetracycline drug. The court held that defendant drug manufacturer failed to establish that civil tort liability was precluded because of an actual conflict with federal law and regulations. It further stated that "Irrespective of what the regulations may have required with regard to changes in labeling, we find no basis for concluding that Lederle was required to continue marketing Declomycin in forms and packaging intended for use by those it believed to be at risk — or indeed to continue marketing at all.” Id. 592 A.2d at 1195. [18] Dr. Barbosa, an American citizen originally from Bolivia, received his medical degree from the University of Madrid, Spain. He completed postgraduate training at Memorial Hospital, Pawtucket, Rhode Island, and Virginia Hospital, Richmond, Virginia. Dr. Barbosa was then assistant professor of pediatrics at the University of Florida’s College of Medicine for three years before taking the position he now holds. He is board-certified in pediatrics and pediatric hematology/oncology. Pl.Ex. 100. [19] Dr. Barbosa had previously testified that heat-treated Factor VIII concentrate killed or inactivated hepatitis and AIDS viruses; that Armour did not sell heat-treated concentrate until May 1984; and that Jason had not previously used heat-treated concentrate because All Children’s Hospital was the first in Florida to use the heat-treated product. After Armour began selling heat-treated Factor VIII concentrate, Dr. Barbosa testified that he thereafter prescribed only heat-treated concentrate. When asked why he switched. Dr. Barbosa stated that there was evidence that the heat-treated concentrate was a safer product. In light of Dr. Barbosa's overall testimony concerning his use of Armour's heat-treated concentrate product, Armour’s statement in its briefs that Dr. Barbosa in fact continued to prescribe Factor VIII concentrate after the AIDS warning was provided is inaccurately misleading. Dr. Barbosa clearly continued to prescribe the heat-treated concentrate, which Armour knows did not transmit the AIDS virus. [20] Other expert medical testimony besides Dr. Barbosa's reviewed various implications of the fact that it had been known before the possibility of HIV-transmission became a concern that the hepatitis virus was transmitted by blood products, including Factor VIII concentrate. Indeed, FDA regulations already in effect in the early 1980’s required manufacturers of Factor VIII concentrate to test each batch of concentrate they produced for the hepatitis virus. If a virus was found, the entire batch of concentrate had to be destroyed. As discussed below, in July 1982, before moving to Florida, Jason had suffered a bout of hepatitis, which began shortly after his receiving three shots of Factor VIII concentrate. [21] Dr. Robinson received his medical degree from the University of Chicago. He completed postgraduate training at Columbia-Presbyterian Medical Center in New York, New York (1960— 1962); at the Department of Biochemistry, University of Chicago (1962-63); at the University of Chicago Hospitals (1963-1964); and at the Virus Laboratory, University of California, Berkeley (1964-1965). Dr. Robinson was then Assistant Professor of Molecular Biology and Research Biologist in the Virus Laboratory, University of California, Berkeley (1965-67); Assistant Professor of Medicine, Division of Infectious Diseases, Stanford University School of Medicine, Stanford, California (1967-1970); Associate Professor of Medicine, Division of Infectious Diseases, Stanford University School of Medicine (1970— 1975); and Professor of Medicine, Division of Infectious Diseases, Stanford University School of Medicine (1975-present). Dr. Robinson’s twenty-page Curriculum Vitae is Pl.Ex. 82. It lists his 119 refereed publications and 66 invited manuscripts published in medical journals and books. It lists his participation on the following editorial boards of medical journals: Infection and Immunity, Journal of Infectious Diseases, Journal of Medical Virology, Journal of Virology, Hepatology, and Virology. It lists his twelve consulting memberships in such organizations as the Scientific Advisory Committee on Microbiology and Virology, American Cancer Society, and the Experimental Virology Study Section, National Institutes of Health. It lists his three awards and eight memberships in societies such as the American Federation for Clinical Research and the American Society for Virology. [22] Dr. Richard Scheaffer received his Ph.D. in Statistics from Florida State University. He was Assistant Professor, Department of Statistics, University of Florida, Gainesville, Florida (1967— 72); Associate Professor, Department of Statistics, University of Florida (1972-77); Professor and Chairman, Department of Statistics, University of Florida (1977-89); and Professor, Department of Statistics, University of Florida (1989-present). Dr. Scheaffer’s nine-page resume is Def. Ex. 12. His papers presented are listed on pp. 2-5; his published articles are listed on pp. 6-8; his published books are listed on p. 8; his contracts and grants, both completed and in effect, are listed on p. 9. His resume lists his membership in five professional and scientific organizations such as the International Statistics Institute. It lists his fifteen instances of professional service such as acting as President of the Florida Chapter of the American Statistical Association and his participation in two advisory boards such as the Mathematics Teacher Education Improve *1504 ment Project. It lists his three honors and his participation in four types of panels and workshops. [23] Harold Kessler, M.D., Armour's expert witness on infectious diseases, stated that he had an honest difference of opinion with Dr. Robinson on certain issues. He also stated that he was not questioning Dr. Robinson's qualifications. In fact, Dr. Kessler stated that he was "not at all” surprised to learn that Dr. Robinson served on the editorial boards of Journal of Medical Virology and Journal of Infectious Diseases, medical journals which had published articles of Dr. Kessler's. TT, Vol. 5, 185-186, 209.