On April 11, 1990, shortly after the birth of her first child, Lisa Nelson, at the age of twenty-five, suffered a debilitating stroke that impaired the left side of her body. At the time of her stroke, Mrs. Nelson had a prescription for, and was using, the drug Parlodel. Approximately five years after suffering the stroke, Mrs. Nelson and her husband, David, (collectively, "the Nelsons") filed this action against Sandoz Pharmaceuticals Corporation ("Sandoz") — the manufacturer of Parlodel. They alleged that the prescription drug had precipitated Mrs. Nelson's stroke. The district court entered summary judgment for Sandoz, concluding that the Nelsons had failed to file their action within the applicable statute of limitations period. The district court also rejected the Nelsons' contention that Sandoz should be equitably estopped from asserting the statute of limitations as a defense to the action because of alleged misrepresentations it made to Mrs. Nelson's physician. For the reasons set forth in the following opinion, we affirm in part and reverse in part the judgment of the district court, and we remand this case for further proceedings.

* BACKGROUND

At a hospital located in Wabash, Indiana, Mrs. Nelson, an Indiana resident, gave birth to her first child on March 21, 1990. In the days following the delivery, a physician prescribed the drug Parlodel for Mrs. Nelson. This medication is designed to suppress lactation in postpartum women. Sandoz Pharmaceuticals, a Delaware corporation, manufactured Parlodel and marketed the product to various individuals, including physicians in Indiana.

On April 5, 1990, Mrs. Nelson began her regimen of Parlodel therapy, which consisted of a 2.5 milligram dose of the medication taken twice a day for fourteen consecutive days. Although her consumption of the medication coincided with the onset of nausea and headaches, Mrs. Nelson continued to follow the prescription. On approximately the seventh day of the Parlodel regimen, Mrs. Nelson collapsed; her husband discovered her lying unconscious on the floor of their home. At a local hospital, physicians treated Mrs. Nelson and concluded that, at the age of twenty-five, she had suffered a severe stroke.

During Mrs. Nelson's hospitalization, Dr. Matthew Sprunger, a treating physician, learned that Mrs. Nelson had taken Parlodel and initially suggested a possible correlation between the medication and his patient's condition. Specifically, in Mrs. Nelson's progress report, Dr. Sprunger noted, "Had been on Parlodel for lactation suppression. Some case reports in the literature of postpartum cardiac and vascular insults in [patients] on Parlodel. [Question or question of] relationship in this case. Will [check] literature."1 Vol. VI, Ex.A. Investigating his theory, Dr. Sprunger turned to the Physicians' Desk Reference ("the PDR") — a medical reference that provides information concerning the uses and side effects of numerous prescription drugs. Under its entry for Parlodel, the PDR indicated that incidents of stroke had been reported among individuals who used the medication. In particular, the reference noted that "[s]erious adverse reactions reported include ... 15 cases of stroke.... An unremitting and progressively severe headache, sometimes accompanied by visual disturbance, often preceded by hours to days, many cases of seizure and/or stroke." Vol. V, R.43, Ex.H. However, the PDR, cautioned that "[t]he relationship of these adverse reactions to Parlodel ... administration has not been established." Id.

Dr. Sprunger next consulted Elaine Thomas, the Sandoz sales representative for the Fort Wayne, Indiana area. Dr. Sprunger informed Ms. Thomas that he sought information from Sandoz concerning a possible link between Parlodel and the occurrence of strokes in postpartum women. During deposition testimony regarding her conversation with Dr. Sprunger, Ms. Thomas stated:

Basically, I told him that we were aware of some episodes of stroke but that they occurred more frequently in the postpartum population than in the normal population anyhow, and it was my understanding that they didn't occur any more frequently with patients treated with Parlodel than they did in the postpartum population.

Vol. VI, Ex.E, at 26. Although Ms. Thomas provided Dr. Sprunger with literature relating to this topic, neither individual recalls the article's author, title or precise contents.

After conducting this investigation, Dr. Sprunger concluded that a connection did not exist between Mrs. Nelson's stroke and Parlodel, and he relayed this determination to Mrs. Nelson's family practitioner. In Mrs. Nelson's progress report, Dr. Sprunger wrote: "Obstetrics, see information on front of chart re: Parlodel and strokes. As per Sandoz information, negative correlation of Parlodel and stroke. Will sign off." Vol. VI, Ex.A. (emphasis in original). During deposition testimony, Mr. Nelson indicated that Dr. Sprunger also relayed these findings to him.

The Nelsons also relate that they asked a second physician, Mrs. Nelson's neurologist, if Parlodel induced the stroke. According to the Nelsons, the neurologist brushed aside this theory. Soon after, Mrs. Nelson's physicians discharged her from the hospital, and she returned home.

In October 1992, Mrs. Nelson gave birth to her second child. After the delivery, Mrs. Nelson informed her nurse that she would not take Parlodel. In particular, on Mrs. Nelson's chart, the nurse wrote "states had `stroke' 3 [weeks] past due last [pregnancy]. States won't take Parlodel." Vol. V, R.43, Ex.J. (emphasis in original). Also in 1992, the Nelsons contacted an attorney and inquired whether Mrs. Nelson's family practitioner bore some responsibility for her earlier stroke. The attorney informed them that they lacked adequate information to proceed with a case against this physician.

Two years later, in August 1994, an investigative news program aired by the National Broadcasting Company ("NBC") concerning the incidence of stroke among postpartum women who had taken Parlodel. Mrs. Nelson viewed this program. Later that year, the FDA issued a notice of its intent to withdraw its approval for the use of Parlodel in the treatment of postpartum lactation.

On December 26, 1995, invoking the diversity jurisdiction of the district court,2 the Nelsons filed this products liability action against Sandoz in the United States District Court for the District of New Jersey, alleging that Parlodel had induced Mrs. Nelson's stroke.3 Faced with a flurry of similar cases, the district court, for purposes of discovery, consolidated the Nelsons' action with those of other Parlodel plaintiffs. When this phase of the litigation had concluded, the Parlodel plaintiffs moved to continue the consolidation of their cases for the purpose of trial; the district court, however, denied the motion. In response to this ruling, Sandoz moved, pursuant to 28 U.S.C. § 1404(a), to transfer each of the actions to the home district of the specific plaintiff who had filed the claim. The district court granted the motion, eventually resulting in the transfer of the Nelsons' case to the United States District Court for the Northern District of Indiana.4

Once before the district court, Sandoz moved for summary judgment, alleging that the Nelsons had failed to file their claims within the applicable statute of limitations period.5 As a threshold matter, Sandoz submitted that Indiana, rather than New Jersey, substantive law governed the litigation. Proceeding on this theory, the pharmaceutical company argued that, under Indiana law, the Nelsons' cause of action accrued in April 1990 — the date of Mrs. Nelson's stroke. According to Sandoz, to have filed a timely claim, the Nelsons should have instituted any action relating to the stroke by April 1992 — the date Indiana's two-year statute of limitations period for product liability suits would have expired on these claims. Although conceding that Indiana recognized the discovery rule,6 Sandoz submitted that the Indiana courts applied the doctrine sparingly and would not permit its invocation in this case. Sandoz further argued that, even if the discovery rule applied, the Nelsons should have known of their claims against the pharmaceutical corporation as early as 1992.

In response, although the Nelsons continued to urge the application of New Jersey substantive law to their claims, they also submitted that, under either New Jersey or Indiana law, they had instituted their suit in a timely manner. Submitting that either state would apply the discovery rule to their case, the Nelsons argued that, under this doctrine, their cause of action accrued on the date of the NBC telecast. The Nelsons submitted that, having learned of their claims only in 1994, they filed this action well within both states' two-year statute of limitations periods for product liability suits. In the alternative, the Nelsons submitted that, based on Sandoz' purported efforts to conceal the link between its medication and strokes, the pharmaceutical company should be equitably estopped from asserting the statute of limitations defense.

After considering the parties' positions, the district court entered summary judgment for Sandoz. The court concluded that the Nelsons failed to file their action within the applicable statute of limitations period. The district court first acknowledged that, because this case had been transferred from the District of New Jersey, that state's choice-of-law rules determined the law, including the applicable statute of limitations, that ought govern the litigation. The district court therefore applied New Jersey's government interest analysis approach to choice-of-law matters to this case. According to the district court, with respect to the discovery rule, a conflict existed between Indiana and New Jersey law. Indiana limited the doctrine to a discrete range of actions in which a "`foreign substance was introduced into the plaintiff's body long before any injury or resultant disease became manifest.'" Mem. Op. at 7 (quoting Covalt v. Carey Canada, Inc., 543 N.E.2d 382, 384 (Ind. 1989)). New Jersey, by contrast, allowed plaintiffs to invoke the discovery rule in all tort cases. Having determined that Indiana and New Jersey had different discovery rules, the district court, relying on New Jersey choice-of-law methodology, set about determining which state had an interest in the application of its own law. It noted that Mrs. Nelson was injured in Indiana, the state in which the Parlodel had been prescribed and ingested. The drug had been manufactured in New Jersey, the headquarters of Sandoz. Turning to the governmental policies of each state, the district court then concluded that the purpose of Indiana's statute of limitations was to "place a temporal limit upon liability for a product's defects." Mem. Op. at 8 (quoting Johnson v. Kemper Indus., Inc., 677 N.E.2d 531, 536 (Ind.Ct.App.1997)). By contrast, continued the court, New Jersey's approach, embodied in its broader discovery rule, "`is to avoid harsh results that otherwise would flow from mechanical application of a statute of limitations.'" Mem. Op. at 8 (quoting Mancuso v. Neckles, 163 N.J. 26, 747 A.2d 255, 256 (N.J. 2000)). This more general rule was designed to prevent unfairness to plaintiffs who might otherwise be time-barred without ever learning that they had a cause of action. Noting that Sandoz' corporate presence in New Jersey was the only relevant contact that the state had with the case, the district court determined that this contact with New Jersey was insufficient to give the state an interest in the application of its own law. The state's contact with the litigation was simply unrelated to the policies of its law. Sandoz' presence in the state has nothing to do, concluded the court, with the purpose of New Jersey's discovery rule which is aimed at providing a plaintiff with additional protection from unfairness. The district court concluded that Indiana's statute of limitations applied to the Nelsons' claims.

Relying on its narrow interpretation of the Indiana discovery rule, the district court then concluded that the Nelsons' action did not fall within the narrow spectrum of cases in which the Indiana discovery rule applied. According to the district court, the Nelsons' cause of action accrued when Mrs. Nelson suffered her injury — 1990. Because the statute of limitations expired in 1992, the Nelsons' claims were time-barred. In addition, the district court rejected the Nelsons' contention that Sandoz should be equitably estopped from asserting the statute of limitations as a defense to their action. In this regard, the district court concluded that the Nelsons had failed to present evidence that Sandoz had concealed intentionally information that was necessary for the Nelsons to bring their claims in a timely fashion.

We review de novo the district court's grant of summary judgment. See Thomas v. Pearle Vision, Inc., 251 F.3d 1132, 1136 (7th Cir.2001). Summary judgment is appropriate "if the pleadings, depositions, answers to interrogatories, and admissions on file, together with affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law." Fed.R.Civ.P. 56(c); see Celotex Corp. v. Catrett, 477 U.S. 317, 322-23, 106 S. Ct. 2548, 91 L. Ed. 2d 265 (1986). Our function is not to weigh the evidence but merely to determine if "there is a genuine issue for trial." Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 249, 106 S. Ct. 2505, 91 L. Ed. 2d 202 (1986). In performing this task, we must construe all facts and draw all reasonable inferences in the light most favorable to the nonmoving party. See id. at 255, 106 S. Ct. 2505.

As a threshold matter, we must resolve the choice-of-law issue in this case — whether Indiana's or New Jersey's statute of limitations and accompanying discovery rule govern the Nelsons' claims. The Nelsons contend that, in conducting its choice-of-law analysis, the district court construed too narrowly the ambit of the Indiana discovery rule, thereby tainting its choice-of-law analysis and leading to the erroneous application of Indiana substantive law to their claims. Although conceding that a misinterpretation of the Indiana discovery rule occurred, Sandoz contends that the district court nevertheless concluded correctly that Indiana law applies to the Nelsons' claims.

To resolve this question, we turn to New Jersey's choice-of-law principles.7 In adjudicating choice-of-law issues, New Jersey employs a two-part governmental interest test that "seeks to apply the law of the state with the greatest interests in governing the specific issue in the underlying litigation." Fu v. Fu, 160 N.J. 108, 733 A.2d 1133, 1138 (N.J.1999). Under this approach, we first must assess whether an actual conflict exists between the competing state laws — "a determination that is made on an issue-by-issue basis." Gantes v. Kason Corp., 145 N.J. 478, 679 A.2d 106, 109 (N.J.1996). If the difference between the state laws is illusory and no conflict exists, our inquiry ends and we apply the law of the forum state. However, if an actual conflict exists, we must move to the second step of the analysis and "identify the governmental policies underlying the law of each state and how those policies are affected by each state's contacts to the litigation and to the parties." Gantes, 679 A.2d at 109. After conducting this inquiry, the court applies to the claims before it the substantive law of the state with greatest interest in the litigation.

Employing the methodology set forth in the New Jersey cases, we first must determine whether an actual conflict exists between the statutes of limitations and accompanying discovery rules of New Jersey and Indiana. As both parties acknowledge, Indiana and New Jersey have identical two-year statutes of limitations for product liability suits. The crucial inquiry, then, is whether the states' discovery rules diverge, thereby creating an actual conflict-of-law.

In many respects, the Indiana and New Jersey discovery rules bear a striking similarity to each other. Certainly, the basic verbal formulations are very similar. Under Indiana law, the discovery rule provides that the statute of limitations begins to run on a cause of action when "the plaintiff knew or, in the exercise of ordinary diligence, could have discovered that an injury had been sustained as a result of the tortious act of another." Horn v. A.O. Smith Corp., 50 F.3d 1365, 1369 (7th Cir.1995); Wehling v. Citizens Nat'l Bank, 586 N.E.2d 840, 843 (Ind. 1992); Doe v. United Methodist Church, 673 N.E.2d 839, 844 (Ind.Ct.App.1996). Differing little in terms of phraseology from its Indiana counterpart, the New Jersey discovery rule states that "a cause of action will be held not to accrue until the injured party discovers, or by the exercise of reasonable diligence and intelligence should have discovered that he may have a basis for an actionable claim." Martinez v. Cooper Hosp.-Univ. Med. Ctr., 163 N.J. 45, 747 A.2d 266, 270 (N.J.2000) (quoting Lopez v. Swyer, 62 N.J. 267, 300 A.2d 563, 565 (N.J.1973)). Moreover, neither state constrains application of its discovery rule to a limited spectrum of cases; rather, a plaintiff may invoke the rule in numerous types of tort actions. Compare Wehling v. Citizens Nat'l Bank, 586 N.E.2d 840, 843 (Ind.1992) (recognizing that Indiana discovery rule applies to all tort claims), with County of Morris v. Fauver, 153 N.J. 80, 707 A.2d 958, 972 (N.J.1998) (listing spectrum of cases in which discovery rule is available). Although the precise events that trigger the statute of limitations for discovery rule purposes will vary from case to case, both states have developed similar principles that guide this fact-intensive inquiry. For instance, as a general rule, neither state requires a plaintiff to know, to a medical or legal certainty, the cause of his injuries before his cause of action will accrue under the discovery rule. Compare Evenson v. Osmose Wood Preserving Co. of Am., Inc., 899 F.2d 701, 705 (7th Cir.1990) ("Events short of a doctor's diagnosis can provide a plaintiff with evidence of a reasonable possibility that another's act or product caused his injuries.") with Lapka v. Porter Hayden Co., 162 N.J. 545, 745 A.2d 525, 530 (N.J.2000) ("We impute discovery if the plaintiff is aware of facts that would alert a reasonable person to the possibility of an actionable claim; medical or legal certainty is not required.").

However, despite these similarities, we may not ignore one pertinent difference between the states' discovery rules. The New Jersey courts consider the discovery doctrine "at its root ... a rule of equity." Lapka, 745 A.2d at 532. More precisely, the New Jersey courts do not limit their inquiry merely to assessing whether a party knew or should have known of his cause of action. Rather, the equitable underpinnings of the discovery rule mandate that the court "consider elements of fairness pertaining to all parties, not just to those asserting the benefit of the [discovery] rule." Id. In essence, New Jersey courts engage in a balancing test that seeks to identify, evaluate and weigh the equitable claims of the opposing parties.8 See Lopez v. Swyer, 62 N.J. 267, 300 A.2d 563, 567 (N.J.1973). Simply put, under New Jersey law, if invocation of the discovery rule would produce an inequity, such as prejudicing the defendant's ability to defend the claim, a plaintiff may not receive the benefit of the doctrine.9 See, e.g., Mancuso v. Neckles, 163 N.J. 26, 747 A.2d 255, 262 (N.J.2000) (considering whether defendant would be prejudiced by application of the discovery rule).

On the basis of our study of the case law of both jurisdictions, we must conclude that an actual conflict exists between New Jersey and Indiana law; the two states employ different analytical approaches to the application of the discovery rule. Before permitting a plaintiff to invoke the discovery rule, New Jersey relies on what it terms equitable principles to assess the prejudice a defendant may incur in defending against a stale claim. New Jersey's use of these equitable considerations differentiates its discovery rule from that of Indiana. Although we acknowledge that the divergence between the two rules is subtle, it is sufficiently significant to be characterized for choice-of-law purposes as an actual conflict.

Confronted with an actual conflict between state laws, New Jersey choice-of-law methodology requires that we now consider each state's interest in enforcement of its rule. See Gantes, 679 A.2d at 109. In performing this task, we must "identify the governmental policies underlying the law of each state and how those policies are affected by each state's contacts to the litigation and to the parties." Veazey v. Doremus, 103 N.J. 244, 510 A.2d 1187, 1189 (N.J.1986). The law of the state with the greatest interest in the litigation will control.

Focusing solely on the issue before us, we conclude that, in this instance, New Jersey would apply the Indiana discovery rule to the Nelsons' claims. The policies underlying the Indiana discovery rule closely align with that state's contact to, and interest in, the litigation and the parties. Specifically, the Indiana discovery rule "is based on the reasoning that it is inconsistent with our system of jurisprudence to require a claimant to bring his cause of action in a limited period in which, even with due diligence, he could not be aware a cause of action exists." UNR-Rohn, Inc. v. Summit Bank of Clinton County, 687 N.E.2d 235, 240 (Ind.Ct.App. 1997) (quoting Barnes v. A.H. Robins Co., Inc., 476 N.E.2d 84, 86 (Ind.1985)). Because the rule seeks to insulate Indiana citizens from the harsh effects of mechanical application of the statute of limitations, Indiana possesses a significant interest in enforcing its discovery rule to its own citizens, including the Nelsons. Moreover, the drug that allegedly caused the injury was administered and consumed within Indiana borders, providing the state with another significant stake in this litigation. Cf. Henry v. Richardson-Merrell, Inc., 508 F.2d 28, 38 (3d Cir.1975). We are mindful that New Jersey's interest in this case is far from negligible. New Jersey holds an interest in deterring tortious conduct of entities such as Sandoz that operate principally within the state's borders. See, e.g., Gantes, 679 A.2d at 115. However, as its state courts note, the "linchpin of the [New Jersey] discovery rule is the unfairness of barring claims of unknowing parties." Caravaggio v. D'Agostini, 166 N.J. 237, 765 A.2d 182, 186 (N.J.2001). The policy underlying the New Jersey discovery rule does not align precisely with that state's interests in this litigation. Accordingly, we conclude that Indiana law — its statute of limitations and accompanying discovery rule — applies to the Nelsons' claims.10

Having concluded that the Indiana discovery rule applies to this case, we must assess whether the Nelsons filed their claim within the applicable statute of limitations period.

As we previously noted, Indiana requires a plaintiff to commence her product liability suit "within two years after the cause of action accrues." Ind.Code § 33-1-1.5-5 (recodified at Ind.Code § 34-20-3-1). However, recognizing ambiguity in the meaning of "accrues," Indiana courts have provided substance to this word through application of the discovery rule. See Degussa Corp. v. Mullens, 744 N.E.2d 407, 410 (Ind.2001). Under Indiana law, the statute of limitations "begins to run from the date that the plaintiff knew or should have discovered that she suffered an injury or impingement, and that it was caused by the product or act of another."11 Degussa Corp., 744 N.E.2d at 410.

As the Indiana courts recognize, the date upon which a plaintiff "discovered facts which, in the exercise of ordinary diligence, should lead to the discovery of [causation] and resulting injury, is often a question of fact." Van Dusen v. Stotts, 712 N.E.2d 491, 499 (Ind.1999). Generally, though, the plaintiff's suspicion, standing alone, about the source of her injury is insufficient to trigger the onset of the limitations period. See Evenson, 899 F.2d at 705; Van Dusen, 712 N.E.2d at 499. In contrast, the limitations period will begin to run when a physician suggests there is a "reasonable possibility, if not a probability" that a specific product caused the plaintiff's injury. Degussa Corp., 744 N.E.2d at 411; Van Dusen, 712 N.E.2d at 499. In this latter case, a reasonable individual, exercising ordinary diligence, would pursue the lead and procure "additional medical or legal advice needed to resolve any remaining uncertainty or confusion regarding the cause of his or her injuries." Degussa Corp., 744 N.E.2d at 411. The Indiana courts have cautioned that "`events short of a doctor's diagnosis can provide a plaintiff with evidence of a reasonable possibility' that another's product caused his or her injuries." Id. at 411 (quoting Evenson, 899 F.2d at 705).

We conclude that, based on the record before us, the district court inappropriately entered summary judgment for Sandoz based on the statute of limitations defense. Although the Nelsons had the burden of establishing, in replying to Sandoz' motion for summary judgment, that the discovery rule ought to apply, they certainly have demonstrated that Sandoz cannot prevail on its motion for summary judgment. Foremost, Sandoz is incorrect when it states that, aside from a single discussion with the neurologist, the Nelsons failed to identify a conversation in which a physician dismissed "the possibility that Parlodel caused her stroke." Appellee's Br. at 10 n. 2. Although Dr. Sprunger initially suggested a possible link between the drug and strokes, he no longer held this tentative view after further inquiry. Indeed, he stated in Mrs. Nelson's medical records that "[a]s per Sandoz information, negative correlation of Parlodel and stroke. Will sign off." Vol. VI, Ex.A (emphasis in original). According to deposition testimony adduced during discovery, Dr. Sprunger relayed these findings to the Nelsons. In particular, Mr. Nelson testified that: "Lisa talked to [Dr.] Sprunger and he said there may be a connection to the drug. And then he — he said there wasn't — when we left, there was no — they didn't know what it was. They had brushed that off, so we didn't think anything else about it." Vol. VI, Ex.D, at 165. According to the Nelsons, when they asked a second physician, the neurologist, about a possible link between the drug and Mrs. Nelson's injuries, he "shoved it off to the side like [there was] no connection whatsoever." Vol. VI, Ex.D, at 157. The neurologist indicated that "he didn't know [the cause] for sure and he said we'll maybe never know for sure." Vol. VI, Ex.D, at 165. Moreover, as we noted earlier, based on information from Sandoz and the PDR, Dr. Sprunger stated in Mrs. Nelson's medical file: "negative correlation of Parlodel and stroke." Vol. VI, Ex.A (emphasis in original). Although Sandoz suggests that the Nelsons should have heeded an earlier note from Dr. Sprunger in Mrs. Nelson's medical records suggesting a link between the drug and strokes, it simply ignores Dr. Sprunger's later entry disavowing this causation theory. Indeed, Sandoz has failed to proffer any reason why this later entry in the medical file would not have been available to the Nelsons.

Beyond the PDR entry for Parlodel and Sandoz' erroneous construction of the record, Sandoz makes no other serious effort to demonstrate that the Nelsons were alerted to a reasonable possibility or probability that Parlodel may have induced Mrs. Nelson's stroke and thus triggered the statute of limitations period prior to 1994. Sandoz does submit that Mrs. Nelson's refusal to take Parlodel in 1992 indicates that she knew or should have known of her potential cause of action against Sandoz at that time. In making medical choices, however, an individual is free to act on mere suspicion or excessive caution; suspicions and hunches do not constitute knowledge concerning causality of injury. See Degussa Corp., 744 N.E.2d at 411.

Finally, we turn to the Nelsons' contention that, under the doctrine of fraudulent concealment, Sandoz should be equitably estopped from asserting the statute of limitations defense in this action. More precisely, the Nelsons contend that Sandoz, through its sales representative Ms. Thomas, provided Dr. Sprunger with false and misleading information designed to forestall their investigation into a possible link between Parlodel and seizures. The district court rejected this position, concluding that the record contained insufficient information to support the allegation of fraudulent concealment. We agree.

Under the doctrine of fraudulent concealment, a defendant is equitably estopped from asserting the statute of limitations defense when, through deception, that party has "concealed from the plaintiff material facts thereby pre venting the plaintiff from discovering a potential cause of action." Fager v. Hundt, 610 N.E.2d 246, 251 (Ind.1993). To invoke the doctrine, the plaintiff must demonstrate that "the wrongdoer was not simply silent but committed affirmative acts designed to conceal the cause of action." Horn, 50 F.3d at 1372 (citations to Indiana law omitted). The affirmative acts must generally rise to the level of some "trick or contrivance intended by the defrauder to exclude suspicion and prevent inquiry." Ludwig v. Ford Motor Co., 510 N.E.2d 691, 697 (Ind. App.Ct.1987). Moreover, the plaintiff must have reasonably relied on the deceptive statements. See Doe v. United Methodist Church, 673 N.E.2d 839, 845 (Ind.Ct.App.1996).

During deposition testimony, Dr. Sprunger stated that he based his conclusion concerning the lack of a link between Parlodel and strokes on "the PDR, and there was an article that was provided to me by the Sandoz rep here in Fort Wayne." Vol. VI, Ex.B, at 9. Neither Dr. Sprunger nor Ms. Thomas, the Sandoz sales representative, is able to recall the article's author, contents or title. Absent this information, it would be impossible for a juror to conclude that the document contained information designed to mislead intentionally Dr. Sprunger and ultimately the Nelsons. Although the Nelsons have contended at various times in this litigation that the article could be identified by a process of elimination, a claim of fraudulent concealment may not rest on such speculative grounds.

In addition, the Nelsons posit that Ms. Thomas, in her conversation with Dr. Sprunger, made statements designed to mislead the physician. In particular, Ms. Thomas stated that:

Basically, I told [Dr. Sprunger] that we were aware of some episodes of stroke but that they occurred more frequently in the postpartum population than in the normal population anyhow, and it was my understanding that they didn't occur any more frequently with patients treated with Parlodel than they did in the postpartum population.

Vol. VI, Ex.E, at 26. The Nelsons contend that an internal memorandum from Sandoz demonstrates the falsity of this assertion. In particular, the memorandum provides that "at this time, [1993,] we have no hard data to demonstrate that there is not an increased risk of stroke ... in Parlodel users." Vol. VII, Ex.821, at 5. Put in other terms, the document simply indicates that the company cannot state definitively whether a link exists between its product and seizures. This statement does not render Ms. Thomas' assertions false or deceptive. Accordingly, the district court properly entered summary judgment on this portion of the Nelsons' claims.

Indiana's statute of limitations and discovery rule govern the Nelsons' claims. We, however, conclude that the district court inappropriately entered summary judgment for Sandoz based on the statute of limitations defense. We agree with the district court that summary judgment remains appropriate on the Nelsons' claim of fraudulent concealment. Accordingly, the judgment of the district court is affirmed in part, reversed in part and the case is remanded for further proceedings. The Nelsons may recover the costs of this appeal.

AFFIRMED IN PART REVERSED IN PART AND REMANDED.