PAUL KELLY, JR., Circuit Judge.
Plaintiffs-Appellants Steven and Marcia Thom (the “Thoms”) appeal from an order granting Defendant-Appellee Bristol-Myers Squibb’s motion for summary judgment. We have jurisdiction over this diversity-based products liability action pursuant to 28 U.S.C. § 1291, and we reverse.
Background
The Thoms allege that Mr. Thom suffered personal injuries as a result of his use of Serzone (nefazodone), a prescription antidepressant medication manufactured by Bristol-Myers Squibb (BMS). According to the Thoms, Mr. Thom’s use of Ser-zone caused him to develop priapism, a “[pjersistent erection of the penis, accompanied by pain and tenderness, resulting from a pathologic condition rather than sexual desire.”
Stedman’s Medical Dictionary
1425 (26th ed.1995). As a result of his priapism, Mr. Thom has permanent penile injury.
Mr. Thom was prescribed Serzone in August 1998 by his physician, Dr. Mark Schueler, to treat his sleep problems and depression. This was the first time that Dr. Schueler had prescribed Serzone. Dr. Schueler learned of Serzone at a hospital presentation that he believes was given by a representative of BMS. Aplt.App. at 324-25. However, he did not stay for the entire presentation.
Id.
at 325.
Serzone contains an FDA-approved package insert, which was reproduced in the Physician’s Desk Reference (PDR), that discusses the possible complications of Serzone use, including priapism. At the time Serzone was prescribed to plaintiff, the package insert read:
PRECAUTIONS
Priapism
While priapism did not occur during pre-marketing experience with nefazodone, rare reports of priapism have been received since market introduction. A causal relationship to nefazodone has not been established (see ADVERSE REACTIONS Section). If patients present with prolonged or inappropriate erections, they should discontinue therapy immediately and consult their physicians. If the condition persists for more than 24 hours, a urologist should be consulted to determine appropriate management.
ADVERSE REACTIONS
Postintroduetion Clinical Experience
Post marketing experience with SER-ZONE has shown an adverse experience profile similar to that seen during the premarketing evaluation of nefazodone. Voluntary reports of adverse events temporally associated with SERZONE have been received since market introduction that are not listed above and for which a causal relationship has not been established. These include:
Rare occurrences of ... priapism (see PRECAUTIONS Section).
As discussed in more detail below, the record is unclear as to whether Dr. Schueler read this package insert prior to prescribing Serzone to Mr. Thom. However, it is uncontested that Dr. Schueler did not discuss the possibility of priapism with Mr. Thom prior to prescribing it. Aplt.App. at 73, 357.
BMS argued it was entitled to summary-judgment because (1) the Thoms could not establish that BMS failed to adequately warn that priapism was associated with Serzone use; and (2) the Thoms could not establish that any failure to warn was the proximate cause of their injuries.
Discussion
A.
Summary Judgment Standard
We review a grant of summary judgment
de novo,
applying the same standard used by the district court under Federal Rule of Civil Procedure 56(c).
Ashley Creek Phosphate Co. v. Chevron, USA, Inc.,
315 F.3d 1245, 1253 (10th Cir.2003).
Summary judgment is proper if the movant demonstrates that there is “no genuine issue as to any material fact” and that it is “entitled to a judgment as a matter of law.” Fed.R.Civ.P. 56(c). In applying this standard, we view the factual record and draw all reasonable inferences therefrom in the light most favorable to the nonmovant.
Matsushita Elec. Indus. Co. v. Zenith Radio Corp.,
475 U.S. 574, 587, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986);
Hirase-Doi v. U.S. West Communications, Inc.,
61 F.3d 777, 781 (10th Cir.1995). An issue is “genuine” if there is sufficient evidence on each side so that a rational trier of fact could resolve the issue either way.
Anderson v. Liberty Lobby, Inc.,
477 U.S. 242, 248, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986). An issue of fact is “material” if under the substantive law it is essential to the proper disposition of the claim.
Id.
In this diversity matter, we must apply the substantive law of the forum state.
Erie R.R. Co. v. Tompkins,
304 U.S. 64, 78, 58 S.Ct. 817, 82 L.Ed. 1188 (1938).
The movant bears the initial burden of making a prima facie demonstration of the absence of a genuine issue of material fact and entitlement to judgment as a matter of law.
See Celotex Corp. v. Catrett,
477 U.S. 317, 322-23, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986). In so doing, a movant that will not bear the burden of persuasion at trial need not negate the nonmovant’s claim.
Id.
at 325, 106 S.Ct. 2548. Such a movant may make its prima facie demonstration simply by pointing out to the court a lack of evidence for the nonmovant on an essential element of the nonmovant’s claim.
Id.
If the movant carries this initial burden, the nonmovant that would bear the burden of persuasion at trial may not simply rest upon its pleadings; the burden shifts to the nonmovant to go beyond the pleadings and “set forth specific facts” that would be admissible in evidence in the event of trial from which a rational trier of fact could find for the nonmovant. Fed.R.Civ.P. 56(e).
B.
The Learned Intermediary Doctrine in Wyoming
BMS contends, and the district court agreed, that as a drug manufacturer and distributor, its duty to warn extends only to physicians. In turn, physicians, based upon knowledge of their own patients, bear the final responsibility for the decision to prescribe medications and to warn the patient of possible side effects. Known as the “learned intermediary doctrine,” this doctrine shields manufacturers of prescription drugs from liability where the manufacturer adequately w^rns a patient’s pre: scribing physician of the potential risks inherent in the use of the product.
Edwards v. Basel Pharms.,
933 P.2d 298, 300 (Okla.1997). The district court predicted that the Wyoming Supreme Court would adopt the learned intermediary doctrine.
The learned intermediary doctrine derives from § 402A of the Restatement (Second) of Torts,
see, e.g., Edwards,
933 P.2d at 300, which the Wyoming Supreme
Court has adopted in its entirety.
Ogle v. Caterpillar Tractor Co.,
716 P.2d 334, 341-42 (Wyo.1986). Section 402A provides that atelier of a defective, unreasonably dangerous product is strictly liable for physical harm to a consumer. Comment k to § 402A establishes that this rule does not apply to “unavoidably unsafe products.” These are
products which, in the present state of human knowledge, are quite incapable of being made safe for their intended and ordinary use. These are especially common in the field of drugs.... Such a product, properly prepared, and accompanied by proper directions and warning, is not defective, nor is it
unreasonably
dangerous.
Restatement (Second) of Torts § 402A cmt. k.
Forty-four other jurisdictions have adopted the learned intermediary doctrine in prescription medicine cases,
see Vitanza v. Upjohn Co.,
257 Conn. 365, 778 A.2d 829, 838 n. 11 (2001) (collecting cases), and the District of Wyoming has predicted that Wyoming would follow suit.
See Jacobs v. Dista Products Co.,
693 F.Supp. 1029, 1036 (D.Wyo.1988) (“The drug manufacturer’s liability to plaintiff ended when it imparted adequate warnings to the physician.”). In addition, the Tenth Circuit has implied in an analogous case that Wyoming would adopt the doctrine.
See Haste v. Am. Home Products Corp.,
577 F.2d 1122, 1125 (10th Cir.1978) (“[T]he defendant discharged its duty to plaintiffs by the warnings to the veterinarians.”).
The Thoms’ main arguments against application of the learned intermediary doctrine are (1) the Wyoming Supreme Court and Legislature have not specifically adopted the doctrine, despite a Wyoming district court’s prediction fourteen years ago that it would; (2) BMS’s provision of a “Patient Information Sheet” to the plaintiff via Dr. Schueler Constitutes a gratuitous undertaking, subjecting BMS to liability; and (3) the policy of Wyoming, in light of the Wyoming comparative fault statute, Wyo. Stat. Ann. § 1-1-109, would not immunize drug manufacturers.
First, silence on the part of the state means only that it has not had occasion to review the matter, not that it disagrees with the federal court’s interpretation of state law. Although the Wyoming Supreme Court has not to date acknowledged the learned intermediary doctrine, “neither has it denied the [doctrine]; it simply has not ruled on the issue. We can and must safely assume that the delay, in the grandest traditions of all common-law courts, is due to the absence of a well presented and soundly argued case, rather than indicative of some invented implication” that the doctrine does not exist.
Rawson v. Sears, Roebuck & Co.,
822 F.2d 908, 927 (10th Cir.1987) (McKay, J., dissenting).
Second, .the “voluntary duty” doctrine is exactly what the learned intermediary doctrine seeks to avoid. Thus, even if a drug manufacturer provides pamphlets for distribution to the ultimate drug user, “[t]he patient is expected to place primary reliance on the physician’s judgment, and to follow his advice and instructions as to use of the drug.”
Seley v. G.D. Searle & Co.,
67 Ohio St.2d 192, 423 N.E.2d 831, 840 (1981);
see also Spychala v. G.D. Searle & Co.,
705 F.Supp. 1024, 1033 (D.N.J.1988);
Wyeth-Ayerst Labs. Co. v. Medrano,
28 S.W.3d 87, 93 (Tex.App.2000);
Presto v. Sandoz Pharms. Corp.,
226 Ga.App. 547, 487 S.E.2d 70, 73-74 (1997).
Third, the Wyoming comparative fault statute has no effect on the application of the learned intermediary doctrine. The learned intermediary doctrine addresses a drug manufacturer’s duty to provide a warning to consumers.
See Wright ex rel. Trust Co. of Kan. v. Abbott Labs., Inc.,
259
F.3d 1226, 1233 (10th Cir.2001). Wyoming’s comparative fault scheme, on the other hand, presents evidence of another’s negligence “in order to reduce damages; [it] in no way defines or affects the scope of the defendant’s initial duty.”
Valance v. VI-Doug, Inc.,
50 P.3d 697, 702 (Wyo. 2002). “The adoption of comparative negligence ... does not abrogate the necessity of an initial finding that the [defendant] owed a duty to [the plaintiff].”
Id.
(internal quotation marks and citations omitted).
C.
Adequacy of Warnings
Physicians become learned intermediaries only when they have received adequate warnings from the drug manufacturer.
See Amore v. G.D. Searle & Co.,
748 F.Supp. 845, 850 (S.D.Fla.1990). Although the adequacy of warnings concerning drugs is generally a question of fact, it can “become a question of law where the warning is accurate, clear and unambiguous.”
Felix v. Hoffmann-LaRoche, Inc.,
540 So.2d 102, 105 (Fla.1989) (collecting cases). An adequate warning of an unapparent risk is one that is reasonable under the circumstances.
Brochu v. Ortho Pharm. Corp.,
642 F.2d 652, 657 (1st Cir.1981). The following considerations, while certainly not exclusive, are relevant in determining whether a warning is adequáte as a matter of law:
1. the warning must adequately indicate the scope of the danger; 2. the warning must reasonably communicate the extent or seriousness of the harm that could result from misuse of the drug; 3. the physical aspects of the warning must be adequate to alert' a reasonably prudent person to the danger; 4. a simple directive warning may be inadequate when it fails to indicate the consequences that might result from failure to follow it and, ... 5. the means to convey the warning must be adequate.
Pittman v. Upjohn Co.,
890 S.W.2d 425, 429 (Tenn.1994) (quoting
Serna v. Roche Labs.,
101 N.M. 522, 684 P.2d 1187, 1189 (1984)).
The district court, citing
Jacobs,
693 F.Supp. at 1032-35, and
Caveny v. CIBA-GEIGY Corp.,
818 F.Supp. 1404, 1406 (D.Colo.1992), reasoned that because the package insert specifically warned of the injury that the plaintiff suffered, it was deemed adequate as a matter of law. D. Ct. Opin. at 10-11. The mere mention of a possible injury, however, is not necessarily adequate. As the Fifth Circuit held in interpreting Louisiana law:
th[e] suggestion that any clear and unambiguous reference to a particular adverse effect is sufficient to- satisfy the manufacturer’s duty to warn is inconsistent with this court’s jurisprudence interpreting the [Louisiana Products Liability Act]. In applying the learned intermediary doctrine to an inadequate warning claim ... a drug manufacturer has discharged its duty to consumers of its prescription drugs when it has
reasonably
informed prescribing physicians of the dangers of harm from such a drug. Thus, ... a mere reference to an adverse effect is not necessarily an “adequate warning.”
Stahl v. Novartis Pharms. Corp.,
283 F.3d 254, 266-67 (5th Cir.2002) (internal quotation marks and citations omitted).
Moreover, the package insert for Ser-zone indicated only that “rare reports” of priapism were “temporally associated” with Serzone; it further stated that a “causal relationship [of priapism] to nefa-zodone has not been established.” Such “warnings” fall well short of those in
Jacobs
and
Caveny,
and we do not think that this equivocal language is adequate as a matter of law.
See Williams v. Lederle Labs.,
591 F.Supp. 381, 385 (S.D.Ohio
1984) (package insert “did not convey a fair indication of the nature of the dangers involved, was reluctant and equivocal in tone and lacked a sense of urgency,” thereby precluding summary judgment).
Compare Amore,
748 F.Supp. at 852 (denying summary judgment where warning stated that incidents of pelvic inflammatory disease were “reported,” and that the reports of such problems were “unconfirmed”),
Salmon v. Parke, Davis & Co.,
520 F.2d 1359, 1363 (4th Cir.1975) (where warning did not unequivocally state that there was a causal relationship between the drug and the side effect, but only that it was “known to occur” after administration, a “jury could reasonably infer that Parke, Davis’ warning lacked the emphasis that the danger demanded”),
and Tongate v. Wyeth Labs.,
220 Ill.App.3d 952, 162 Ill.Dec. 801, 580 N.E.2d 1220, 1224 (1999) (denying summary judgment where warnings noted that injury was “reported” and “temporally associated” with the use of the drug, and that a “causal relationship has not been established”),
with Jacobs,
693 F.Supp. at 1032 (granting summary judgment where warning stated potentially “life-threatening” side effect was reported and indicated the primary cause and management of it),
and Cavervy,
818 F.Supp. at 1405 (granting summary judgment where package insert noted that a “causal relationship [was] probable”).
In determining the adequacy of a warning, a court must also look to evidence concerning the manufacturer’s knowledge of the danger of the product.
See, e.g., Ross v. Jacobs,
684 P.2d 1211, 1215 (Okla.Civ.App.1984). Because, the duty to warn arises only when the manufacturer knows or should know of the risk, the adequacy of a warning is commensurate with the manufacturer’s knowledge of the effects of the drug.
See Basko v. Sterling Drug, Inc.,
416 F.2d 417, 426 (2d Cir.1969);
see also Jacobs,
693 F.Supp. at 1034 (“The state of medical knowledge at that time will directly affect a decision on the adequacy of defendants’ warnings.”).
In this case, there is a factual controversy regarding the state of BMS’s knowledge of priapism and the relation between nefazodone (Serzone) and trazodone, another drug manufactured by BMS under the brand name Desyrel. The Thoms have produced evidence that during the testing of nefazodone, BMS referred to nefazodone as the “chemical and pharmacologic analog of trazodone.” Aplt.App. at 130. Trazodone was introduced into the marketplace in 1982, some thirteen years before nefazodone was introduced into the United States market. Within .months of trazodone’s introduction, cases of priapism associated with the use of trazodone began being reported. ApltApp. at 101. A little over a year after trazodone was introduced into the marketplace, published peer review medical articles began reporting the association between trazodone and priapism.
See, e.g.,
Gerald M. Aronoff,
Trazodone Associated with Priapism,
1984:1 Lancet 856; Maryonda Scher et al.,
Trazodone and Priapism,
140 Am. J. of Psychiatry 1362 (1983). Against this backdrop of knowledge possessed by BMS regarding the relation between nefazodone and trazodone, coupled with trazodone’s clear association with priapism, the Thoms have presented a genuine issue as to the adequacy of the Serzone package insert warning.
Because of this, “[i]t is for the jury to determine whether defendants possessed actual or constructive knowledge of [the drug’s] potential to produce harmful side effects.”
Bikowicz v. Nedco Pharm., Inc.,
130 A.D.2d 89, 517 N.Y.S.2d 829, 832 (1987).
Finally, there is also a dispute as to whether the instruction to seek medical treatment accurately conveyed the magnitude of the harm, and whether it caused confusion when following the instruction to discontinue treatment. The package insert stated: “If patients present with prolonged or inappropriate erections, they should discontinue therapy immediately and consult their physicians. If the condition persists for more than
21 hours,
a urologist should be consulted to determine appropriate management.” (emphasis added). Dr. Irwin Goldstein, the Thoms’ urological expert, stated in his affidavit that low-flow priapism is a medical emergency that requires medical intervention and resolution within four to eight hours of the onset in order to prevent permanent injury. Aplt.App. at 97. Additionally, the Thoms have produced evidence tending to show that the state of the medical knowledge is that priapism linked with prescription drug use can result in permanent penile injury and impotence if not treated within four to eight hours after its onset. Aplt.App. at 55 (citing Michael T. Compton et ah, Priapism
Associated with Conventional and Atypical Antipsychotic Medications: A Review,
62 J. Clinical Psychiatry 362 (2001); A. Melman
&
S. Serels,
Priapism,
12 Suppl. (4) Int’l J. Impotence Res. S133 (2000)).
In addition to a genuine issue as to the accuracy of the 24 hour instruction, there is also a question of whether the instruction was confusing and ambiguous when compared to the immediately preceding sentence that states that “[i]f patients present with prolonged or inappropriate erections, they should discontinue therapy immediately and consult their physician.” These questions preclude a summary judgment for BMS.
D.
Proximate Cause
Ordinarily what constitutes proximate cause is a question of fact, but the issue of proximate cause becomes a question of law when the facts are undisputed and there is no evidence from which a reasonable trier of fact could find a causal connection between the allegedly negligent act and the injury.
Eck, v. Parke, Davis & Co.,
256 F.3d 1013, 1017 (10th Cir.2001). The vast majority of jurisdictions hold that where a warning is inadequate, the plaintiff is entitled to a rebuttable presumption that an adequate warning would have been heeded if one had been given.
See, e.g., Daniel v. Ben E. Keith Co.,
97 F.3d 1329, 1333 (10th Cir.1996);
Garside v. Oseo Drug, Inc.,
976 F.2d 77, 81 (1st Cir.1992);
Snawder v. Cohen,
749 F.Supp. 1473, 1479 (W.D.Ky.
1990);
see also
Restatement (Second) of Torts § 402A cmt. j.
However, the defendant can rebut the presumption through testimony that a different warning would not have made a difference in the actions of the physician.
Garside,
976 F.2d at 81. If the manufacturer rebuts the presumption of causation, then the plaintiff must further establish proximate cause by showing that, had the manufacturer issued a proper warning to the physician, he would have altered his behavior and injury would have been avoided.
Miller v. Pfizer Inc.,
196 F.Supp.2d 1095, 1127 (D.Kan.2002). To create a factual question, the evidence must be of sufficient weight to establish, by the preponderance of the evidence, at least some reasonable likelihood that an adequate warning would have prevented the plaintiff from receiving the drug.
See Thomas v. Hoffman-LaRoche, Inc.,
949 F.2d 806, 812 (5th Cir.1992);
Brochu v. Ortho Pharm. Corp.,
642 F.2d 652, 660 (1st. Cir.1981).
The test for proximate cause in a pharmaceutical failure to warn case is whether the defendant’s inadequate warning could be found to be a substantial cause of the plaintiffs ingestion of the drug.
McEwen v. Ortho Pharm. Corp.,
270 Or. 375, 528 P.2d 522, 538 (1974);
see also Buckley v. Bell,
703 P.2d 1089, 1091 (Wyo.1985). The majority of courts that have examined the issue have held that when a physician fails to read or rely on a drug manufacturer’s warnings, such failure constitutes the “intervening, independent and sole proximate cause” of the plaintiffs injuries,
even where the drug manufacturer’s warnings were inadequate. Formella v. Ciba-Geigy Corp.,
100 Mich.App. 649, 300 N.W.2d 356, 358-59 (1980) (“[T]he fact [that] Dr. Murguz failed to read the package inserts and PDR negates any possible negligence on the part of Ciba-Geigy in not emphasizing the hazards in those publications.”);
see also Harris v. McNeil Pharm.,
No. CIV 3:98CV105, 2000 WL 33339657, at *3 n. 3 (D.N.D. Sept. 5, 2000) (“The presumption that had an adequate warning been given it would have been read and heeded is rebutted by [the physician’s] testimony that he did not read the warning.”);
Oppenheimer v. Sterling Drug, Inc.,
7 Ohio App.2d 103, 219 N.E.2d 54, 58 (1964) (“Even assuming negligence on the part of the defendant ... there is nothing to indicate that the doctor relied upon any information furnished by the defendant in prescribing Aralen for his patient, the plaintiff....”);
Douglas v. Bussabarger,
73 Wash.2d 476, 438 P.2d 829 (1968) (“[I]f defendant-drug company was negligent in not labeling its container so as to warn of dangers, this negligence was not a proximate cause of plaintiffs disability” because plaintiffs doctor “did not read the labeling which was on the container.”).
The district court held that because “the uncontroverted facts demonstrated] that Dr. Schueler did not read, nor rely upon, the Serzone package insert or the PDR, and as a result, failed to consider instructing Mr. Thom about the possible occurrence of priapism,” the Thoms were unable to establish that a different warning would have avoided the plaintiffs injuries. D.Ct. Opin. at 15. However, we do not read the record as indicating so unequivocally that Dr. Schueler failed to read the package insert.
BMS relies on the following deposition excerpts by Dr. Schueler to establish that he did not read the Serzone package insert before prescribing it to Mr. Thom.
Q. Do you know if you read the PDR entry for Serzone before you prescribed it for Mr. Thom?
A. I don’t know.
ApltApp. at 336.
Q. Have you ever recalled seeing that entry before November of 1998?
A. No.
Id.
at 357.
Q. Do you know if you ever looked up Serzone in that reference?
A. I probably did.
Q. Any specific recollection
of
doing that?
A. No....
Id.
at 341. In addition, BMS points to the following excerpt from Dr. Schueler’s affidavit to show that he did not read the package insert before prescribing Serzone to Mr. Thom:
8.
After November 1998,
I reviewed the June 1997 BMS Serzone package insert regarding priapism.
Id.
at 73 (emphasis added).
At most, these quoted excerpts demonstrate that Dr. Schueler could not remember, three years after the fact, if he had read and reviewed the complete Serzone package insert with Mr. Thom when he first prescribed the drug to him. They do not support the summary judgment conclusion of BMS and the district court that Dr. Schueler never read or relied upon the package insert or the PDR when he discussed Serzone and its side effects with Mr. Thom.
In fact, various excerpts of Dr. Schueler’s deposition, when viewed in the light most favorable to the Thoms, indicate that Dr. Schueler may have read the package insert or the PDR prior to prescribing Serzone to Mr. Thom:
Q. Do you know if, at or around that time, you undertook to read the package insert for Serzone?
A. I read parts of it, if not all of it. And I couldn’t say exactly when.
Q. Do you know if it was at or around the time of this meeting, presentation [in mid 1998]?
A. I believe it was.
Id.
at 326-27.
Q. You knew at the time you prescribed Serzone that Mr. Thom was on Ritalin?
A. I did.
Q. Did you do any research to determine whether there might be any drug interactions implicated by your prescription?
A. I believe I did.
Q. And what did you do?
A. I don’t have a specific memory, but using one of the references in the exam room, looked up any specific interactions.
Q. What references were available to you in the exam room at that time?
A. The Physicians’ Desk Reference and the monthly prescribing bulletin.
Id.
at 340.
Q. Did you ever review with Mr. Thom the full package insert?
A. As best I remember, we went through the most frequent side effects of the drug. And that could have been the PDR reference....
Id.
at 334.
BMS has not shown that there is a lack of a genuine issue of material fact with regard to whether Dr. Schueler read the package insert.
See Celotex,
477 U.S. at 325,106 S.Ct. 2548. Indeed, it seems to us that BMS has pointed out the presence of a genuine issue, as BMS has admitted in its summary judgment papers that “[w]hether Dr. Schueler ever read the entire package insert for Serzone, and if so, when, remains unclear.” ApltApp. at 32. Under the proper summary judgment framework, the inferences to be drawn from this ambiguity must be resolved in
favor of the Thoms. The evidence is simply not uncontroverted on this point.
REVERSED. All outstanding motions are denied.