Carol Lee Stallings brought this diversity action on behalf of herself and the estate of her late husband, seeking damages from the Georgia-Pacific Corporation for the asbestos-related illness that ultimately took his life. After enduring near-constant exposure to asbestos for some four years in the U.S. Navy, Mr. Stallings worked for several more years with a Georgia-Pacific product containing asbestos at his job finishing drywall. Neither of the medical experts that Stallings consulted for this suit, however, could quantify the extent of her husband’s exposure resulting from his contact with Georgia-Pacific’s products, and instead testified that any further exposure would have contributed to the development of his disease. Heeding this court’s decisions rejecting evidence of that kind as too insubstantial under Kentucky’s standard for causation, the district court accordingly granted summary judgment for Georgia-Pacific, and Stallings now appeals. Because this court’s precedents clearly foreclose the theory of causation on which Stallings relies against Georgia-Pacific, and the Kentucky Supreme Court has not clearly indicated that it would rule otherwise, the district court’s grant of summary judgment was proper.

During his nearly four years in the Navy, William Stallings served aboard the destroyer USS Waller, helping to operate and maintain the ship’s boilers. In that job he worked daily on and amid the ship’s *550valves, pumps, piping, boilers, and turbines, all the while breathing in the asbestos-laced dust that their insulation and other coating materials gave off. The resulting asbestos exposure that he endured was, by all accounts, considerable.

After leaving the service Stallings went to work as a drywall finisher, first for Timmerman Drywall for a year, and then for Stigler Drywall for another two. In those jobs, too, Stallings regularly encountered asbestos-laden materials, this time in the form of the “mud” that he and his fellow finishers used to paste together drywall panels and which they later sanded down, stirring up asbestos-tainted dust. Stallings would later identify one of the materials he used on those jobs—and whose asbestos-laced dust he inhaled multiple times a week—as Bestwall, a product of the Georgia-Pacific Corporation. Several years later, while finishing two rooms in his home, Stallings again used a Georgia-Pacific drywall mix containing asbestos, and for two weeks or so inhaled the dust that the dried mixture unleashed when sanded.

In September 2011, Stallings received a diagnosis of mesothelioma, an incurable cancer resulting from exposure to asbestos. A year later he filed suit in Kentucky state court against Georgia-Pacific and the other manufacturers of the asbestos-containing products he had .been exposed to decades earlier, seeking punitive damages under theories of strict liability and negligence. His wife, Carol Lee, also claimed damages for loss of consortium. The case was later removed to the federal district court, and proceeded there until September 2013, when Stallings died of complications related to his mesothelioma. Soon after, Carol Lee Stallings filed an amended complaint as the surviving spouse and as executrix of Mr. Stallings’ estate, and added a wrongful-death claim.

The remaining defendants, including Georgia-Pacific, eventually moved for summary judgment, which the district court granted. Stallings v. Georgia-Corp., 2015 WL 7258518, at *2 (W.D. Ky. Nov. 17, 2015). As to the claims against Georgia-Pacific, the court found that Stallings had failed to establish that the company’s products were a substantial factor in bringing about Mr. Stallings’ cancer, as required for a finding of causation under Kentucky common law. Id. at *6. Pointing to this court’s line of cases-beginning with Martin v. Cincinnati Gas & Electric Co., 561 F.3d 439 (6th Cir. 2009), the district court noted that in order for the claims against Georgia-Pacific to survive a motion for summary judgment, Stallings would have had to provide evidence that the company’s products were probably, rather than possibly, a “substantial cause” of her husband’s mesothelioma. Stallings, 2015 WL 7258518, at *6. But Stallings’ medical experts could testify only “that any exposure to asbestos qualifies as a substantial exposure,” offering no more precise an estimate of how much of that exposure was due specifically to Georgia-Pacific’s products. Id. Determining that this “any exposure” theory of causation had been foreclosed by Martin and its progeny, the court concluded that Stallings therefore could not convince a “reasonable jury that [Mr. Stallings’] exposure to Bestwall was a probable cause of his mesothelioma,” especially not “in light of his substantial prior asbestos exposure while in the Navy.” Id. The district court accordingly dismissed the claims against Georgia-Pacific, and Stallings now appeals solely from that dismissal.

The district court’s grant of summary judgment was proper because there was not enough evidence of Stallings’ exposure to Georgia-Pacific’s asbestos-containing products to establish them as a likely *551cause of the cancer that took her husband’s life. Under the Kentucky law that governs this diversity action, see Martin, 561 F.3d at 442, Stallings must show that Georgia-Pacific’s products were a substantial factor in bringing about Mr. Stallings’ disease rather than just a factor, see Moeller v. Garlock Sealing Techs., LLC, 660 F.3d 950, 953-54 (6th Cir. 2011) (citing Deutsch v. Shein, 597 S.W.2d 141, 144 (Ky. 1980)). Because this court has already rejected the type of evidence on which Stall-ings relies as simply too insubstantial to satisfy that standard, Stallings cannot show that Georgia-Pacific was legally responsible for Mr. Stallings’ injury.

Our precedents foreclose the theory at the heart of Stallings’ case for causation: that any further exposure to asbestos would have been a substantial factor in bringing about an asbestos-related disease like mesothelioma. As we held in Moeller and Martin, this “any” or “every exposure” theory of causation cannot satisfy Kentucky’s “substantial factor” standard, as that theory would “make every incidental exposure to asbestos a substantial factor,” rendering that standard, and its substantiality requirement, all but “meaningless.” Martin, 561 F.3d at 443 (citation omitted). Stallings nevertheless appears to rely on exactly that sort of theory in making her case for causation against Georgia-Pacific. Neither of her two medical experts could quantify the extent of Mr. Stallings’ exposure to the asbestos in the company’s products, instead insisting that every further exposure would have contributed to the development of hi^ disease.1 But testimony of that kind—the only that Stallings can cite on behalf of her claim that Georgia-Pacific’s products were legally responsible for her husband’s cancer—is too spare to satisfy Kentucky’s “substantial factor” test, especially given the evidence of Mr. Stallings’ considerable daily exposure to asbestos aboard the USS Waller. See Moeller, 660 F.3d at 955, Stallings has accordingly failed to present evidence showing that those products were a probable, as opposed to a merely possible, cause of Mr. Stallings’ disease. Under Kentucky law, as this court has understood it, that failure is not enough to make a reasonable inference of causation, see Martin, 561 F.3d at 443 (citing Bailey v. N. Am. Refractories Co., 95 S.W.3d 868, 873 (Ky. Ct. App. 2001)). Georgia-Pacific was therefore entitled to summary judgment.

This conclusion, moreover, is not undermined by the Kentucky Supreme Court’s decision in CertainTeed Corp. v. Dexter, 330 S.W.3d 64 (Ky. 2010), a case never mentioned by either party until this court requested supplemental briefing to address it. CertainTeed came down a year after this court decided Martin, which, as explained, squarely rejected the “every exposure” theory as inconsistent with Kentucky’s “substantial factor” test for causation. Just a year after CertainTeed was decided, this court once again applied Martin’s holding in Moeller, without considering or indeed even mentioning the intervening ruling in CertainTeed and what impact, if any, it had on Martin’s *552rejection of the “every exposure” theory. Our published reaffirmance in Moeller of Martin would therefore raise a difficult issue regarding stare decisis if the Kentucky Supreme Court had squarely adopted the “every exposure” theory in CertainTeed. However, it is far from clear that the CertainTeed court actually endorsed that theory.

CertainTeed involved a trial court’s grant of a new trial for several corporate defendants whose products, much as in this case, had allegedly exposed the plaintiff Dexter to asbestos. In the original trial the jury had declined to apportion any fault to several other empty-chair defendants, while in the second trial ordered by the trial court substantial percentages of fault were assigned to those defendants. The intermediate appellate court reinstated the original verdict declining to apportion fault to the empty-chair defendants, and the defendants obtained review in the Kentucky Supreme Court.

The Kentucky Supreme Court noted at the outset that:

[No] party in the present appeal challenged the appropriateness of the trial court’s approach to the strict liability and negligence claims against Certain-Teed and the other defendants, though there may be questions about whether both types of claims can be pursued against any single defendant. These issues, and others, may have been the subject of the appeal and cross-appeal of the second trial but were not addressed because of the Court of Appeals’ resolution of reinstating the original judgment. Such issues are not currently before this Court.

CertainTeed, 330 S.W.3d at 70-71. The court went on to examine “certain subsidiary issues,” but:

only to the extent necessary to resolve the primary issue of the appropriateness of the trial court’s grant of a new trial. For the most part, this opinion addresses those subsidiary issues as they have been framed by the parties, without further complication by other issues that might be unresolved, either because they have not been raised or have not yet been addressed by the Court of Appeals.

Id. at 71.

With this highly limiting qualification in place, the court proceeded to explain that “[e]mpty-chair defendants who have settled are to be treated no differently than participating defendants in regard to what must be proved to apportion fault against them.” Id. at 73-74. Then, applying a deferential “clearly erroneous” standard of review to the trial judge’s determination to order a new trial, the Supreme Court found sufficient evidence to support the grant of a new trial.

In conducting that deferential review, the Supreme Court detailed “[a] plethora of evidence show[ing] that Dexter was exposed to asbestos by many of the empty-chair defendants,” id. at 75, and that with respect to causation, “there was other, more specific evidence of legal causation,” id. at 77.

To be sure, the Supreme Court relied “first and foremost” on the testimony of two doctors—one of Dexter’s own experts at trial as well as his treating physician— who advanced an “every exposure” theory against the empty-chair defendants. See 330 S.W.3d at 78. The court, however, also relied on the testimony of another of Dexter’s ’ medical experts, Dr. Hammar. He had explained at trial—apparently at odds with the “every exposure” theory—that whether a product containing asbestos could be said to cause a related injury depended on “the intensity, the duration, and the number of years or months or *553whatever they were exposed to it,” id. at 78 (internal quotation marks omitted). Although the court noted that this testimony did not necessarily conflict with the empty-chair defendants’ liability, as Dr. Hammar had set no “minimum cut-off of exposure,” id. the court nevertheless did not expressly rely on that point to reach its ultimate conclusion about causation—that the trial court had not clearly erred by “finding it to have been unreasonable for the jury to conclude that none of the empty-chair defendants contributed at. all to Mr. Dexter’s disease,” id. That conclusion was instead premised on what the court characterized as “prior evidence of these factors and the extremely high concentration of asbestos fibers in Dexter’s lungs (Dr. Hammar described it as the most he had ever seen in a pipe fitter’s lungs).” Id. Thus, the “ample evidence” that supported a finding of causation need not have been the expert testimony as to the “every exposure” theory. It could equally well have been the evidence that the court explicitly cited when drawing its conclusion on causation: evidence speaking to the intensity and duration of Dexter’s exposure to the defendants’ asbestos-laden products and the quantity of asbestos later found in his lungs.

In addition, the Supreme Court was “further compelled” by the plaintiffs’ admissions in their opening statement. In that statement, Dexter’s attorney appeared to concede that, based on the evidence of Dexter’s exposure to their products, the empty-chair defendants were also at fault for his injuries, to no small degree:

We’re not trying to suggest that GE or Johns-Manville or some other company didn’t have a role or responsibility. No, we think that there’s many companies that participated in causing the death of Mr. Dexter. We think these companies are significant and the evidence mil show that they caused Mr. Dexter to have significant exposure to their products.

Id. at 78 (emphasis added). Although the CertainTeed court declined to rely solely on this “factual admission” in affirming the trial court, it nevertheless made clear that it saw that statement as having “no doubt colored the trial court’s perception of the evidence, putting it on notice that the plaintiff may have been trying to have it both ways.” Id. at 78 n.8.

Thus the Kentucky Supreme Court at most determined that, under a clearly erroneous scope of review, the district court had found enough evidence of causation to get to a jury in a case where two doctors relied on an “every exposure” theory, but another doctor testified in favor of causation without relying on such a theory, and plaintiffs’ original opening statement—asserting causation—“further compelled” deference to the trial court’s determination. This analysis hardly requires the conclusion that “every exposure” expert testimony is by itself sufficient to get to a jury.

This is perhaps why we took no notice of CertainTeed in once again rejecting that theory in Moeller, and why the parties in that case, like those now before the court, did not even think to raise CertainTeed themselves,2 In any event, it is clear enough that the CertainTeed court has not unequivocally resolved the issue. In the absence of clearer ruling from Kentucky’s highest court, keeping in mind this court’s duty to “attempt to ascertain how that court would rule if it were faced with the issue,” Meridian Mut. Ins. Co. v. Kellman, 197 F.3d 1178, 1181 (6th Cir. 1999), the fairer conclusion is that the evidence does *554not yet definitively indicate that court would embrace the “every exposure” theory. Until Kentucky’s courts have more clearly signaled their readiness to accept that theory, we are bound to reject it here once again, in line with this court’s decisions in Moeller and Martin.

The judgment of the district court is affirmed.

Circuit Judge, dissenting.

I disagree with the majority that the “any” or “every exposure” theory of causation is foreclosed by our holding in Martin v. Cincinnati Gas & Electric Co., 561 F.3d 439 (6th Cir. 2009), and its progeny. Although we are generally bound by the principles of stare decisis, our decisions in diversity actions are controlling only insofar as the state supreme court remains silent on the matter. Where, as here, a state supreme court later clarifies the state law, we must apply the law in accordance with those decisions. A year after Martin, the Kentucky Supreme Court made clear in CertainTeed Corp. v. Dexter, 330 S.W.3d 64 (Ky. 2010), that the “any” or “every exposure” theory is an acceptable theory of causation that can satisfy Kentucky’s “substantial factor” standard. Because I believe that CertainTeed controls this case, I respectfully dissent.

As a federal court sitting in diversity, we are obliged to “apply state law in accordance with the controlling decisions of the state supreme court.” Allstate Ins. Co. v. Thrifty Rent-A-Car Sys., Inc., 249 F.3d 450, 454 (6th Cir. 2001); see also Erie R.R. Co. v. Tompkins, 304 U.S. 64, 78, 58 S.Ct. 817, 82 L.Ed. 1188 (1938). Although it is our duty, where the state’s highest court is silent, to “ascertain from all available data what the state law is,” Bailey v. V&O Press Co., Inc., 770 F.2d 601, 604 (6th Cir. 1985), where the state supreme court subsequently issues a decision contrary to our own, we are required to adopt prospectively the state decision clarifying state law. See Vandenbark v. Owens-Illinois Glass Co., 311 U.S. 538, 543, 61 S.Ct. 347, 85 L.Ed. 327 (1941) (“appellate tribunals ... should conform their orders to the state law as of the time of the entry”); Rutherford v. Columbia Gas, 575 F.3d 616, 619 (6th Cir. 2009).

In Martin, we held that the theory on which Stallings relies, that “any” or “every exposure” is a substantial cause of his asbestos-related injury, was insufficient to prove causation under Kentucky law, because “it would render the substantial factor test ‘meaningless.’ ” Martin, 561 F.3d at 443-44. Subsequent to our holding in Martin, however, the Kentucky Supreme Court in CertainTeed relied on the same “every exposure” theory to support a finding of causation. CertainTeed, 330 S.W.3d at 78. Specifically, in assessing the liability of certain empty-chair defendants who manufactured asbestos-containing products, the CertainTeed court held that “the primary evidence of causation in this case was from the medical experts” who testified that “every single exposure to asbestos would have been the legal cause of [plaintiffs] illnesses.” Id. at 77-78. Testimony from two experts who advanced this theory, and one other expert whose testimony was consistent with that theory, provided “ample evidence” that exposure to the empty-chair defendants’ products was a substantial factor in bringing about plaintiffs injury. Id. The Kentucky Supreme Court concluded that “there was uncontroverted evidence that each exposure to asbestos would have been a legal cause of [plaintiffs] injuries. Consequently, the evidence of exposure to the empty-chair defendants’ products means that they must have legally caused some por*555tion of [plaintiffs] injuries.” Id. at 78-79 (emphasis added).

Rather than follow the Kentucky Supreme Court’s decision, the majority attempts to distinguish CertainTeed, arguing that the state court did not squarely adopt the “every exposure” theory. They point, first, to the limiting language contained in the opinion. That language, however, has no bearing on the question of legal causation, or the appropriateness of the “every exposure” theory in proving causation. The state supreme court declined to review wholly separate issues that had been the subject of prior proceedings, including the scope of the second trial, and whether plaintiff could bring both strict liability and negligence claims against each defendant.1 Id. at 70-71. After outlining what questions were not at issue, the Certain-Teed court went on to address the questions that were, including “what a defendant must prove to obtain apportionment against an empty-chair defendant,” Id. at 71. Even under a deferential “clearly erroneous” standard, this “require[d] that there was evidence sufficient to prove fault on the part of at least some of the empty-chair defendants.” Id. at 74. The state supreme court opinion makes clear that this holding required evidence of legal causation. Id. at 77. The acceptability of the “every exposure” theory fits squarely within the scope of the narrow question before the Kentucky Supreme Court. Under these circumstances, we are not entitled to disregard the decision of the state supreme court on the basis of contradictory precedent from our circuit.

The majority next contends that the state court’s adoption of the “every exposure” theory is not binding because the evidence supporting a finding of causation in CertainTeed “could equally well have been” additional evidence cited by the court. The majority mischaracterizes the CertainTeed court’s treatment of this additional evidence. None of the additional evidence was alone sufficient to support the court’s ultimate conclusion that there was evidence of causation. The Kentucky Supreme Court explicitly rejected the argument, advanced here by the majority, that Dr. Hammar’s testimony contradicted the “every exposure” theory. Id. at 78. It noted, instead, that Dr. Hammar’s testimony was “consistent with” the testimony of plaintiffs’ expert and treating physician, and that a contrary interpretation “would only work if Dr. Hammar had set a minimum cut-off of exposure.” Id. In other words, the testimony he provided regarding the intensity and duration of exposure was not dispositive unless he also established the level at which exposure could be classified a substantial factor. In addition, the admission made by plaintiff’s counsel in their opening statement was not sufficient to prove causation. In fact, the Kentucky Supreme Court specifically declined to affirm the trial court "solely on the basis of counsel’s opening statement.” Id. at 78 n.8. None of the additional evidence was sufficient to support a finding of substantial causation.

In contrast, adoption of the “every exposure” theory was necessary to the outcome in CertainTeed. Even if the court did rely on additional evidence such as the “high concentration of asbestos fibers in [plaintiffs] lungs,” there was no other testimony about what level of exposure was sufficient to cause the injury. Id. at 78. The Kentucky Supreme Court made clear that evidence of exposure was insufficient to prove causation; it required a showing that the exposure was a substantial factor in bringing about the injury. Id. at 77. The court *556relied upon the “every exposure” theory to determine the level at which exposure is considered a substantial factor. Id. at 78. The conclusion reached by the Kentucky Supreme Court is simply not possible unless the court squarely adopted the “every exposure” theory as an acceptable theory of causation.

As the majority notes, one year after CertainTeed was decided, we reaffirmed Martin’s rejection of the “every exposure” theory without any reference to Certain-Teed. See generally Moeller v. Garlock Sealing Tech., LLC, 660 F.3d 950 (6th Cir. 2011). The majority argues that Moeller did not address CertainTeed because the Kentucky Supreme Court’s limited holding did not alter our prior rejection of the “every exposure” theory. Yet the majority admits that Moeller failed to consider or even mention the CertainTeed case. There is nothing to suggest that the failure to mention the case in Moeller, which was issued just a year after CertainTeed, was anything more than an oversight. Our duty to adhere to the decisions of state supreme courts in diversity cases may not be evaded merely because our cases mistakenly overlook prior state court decisions. See United States v. Maness, 23 F.3d 1006, 1009 (6th Cir. 1994) (disregarding a Fourth Circuit case that failed to follow a contrary prior state supreme court decision); Factors Etc., Inc. v. Pro Arts, Inc., 652 F.2d 278, 283 (2d Cir. 1981) (“A federal court ... would be obliged to disregard a state law holding by the [federal] court of appeals if persuaded ... that prior state court decisions had been inadvertently overlooked by the pertinent court of appeals”).

“Where a state’s highest court has spoken to an issue, we are bound by that decision unless we are convinced that the high court would overrule it if confronted with facts similar to those before us.” Kurczi v. Eli Lilly & Co., 113 F.3d 1426, 1429 (6th Cir. 1997) (quoting Kirk v. Hanes Corp., 16 F.3d 705, 707 (6th Cir. 1994)). I find no justification for straying from CertainTeed on the basis of the facts in this case. In fact, the testimony provided by Stallings’s experts is indistinguishable from the medical testimony on which the CertainTeed court relied. Stallings’s expert, Dr. Arthur L. Frank, who also testified on behalf of the plaintiff in Cer-tainTeed, testified that “[t]he cumulative exposures that [Stallings] had to asbestos from any and all products containing any and all fiber types would have contributed to his developing this malignancy.”2 R. 172-4 (Frank Dep. at 43-44) (Page ID #5642). Stallings’s other medical expert, Dr. James Strauchen, expressed a similar view. He stated that “[a]ll of [Stallings’s] exposures to asbestos would have been substantial causes of his mesothelioma.” R. 172-6 (Strauchen Dep. at 21) (Page ID #5703).

In declining to adopt the “every exposure” theory, the majority makes the odd assertion that

the Kentucky Supreme Court at most determined that, under a clearly erroneous scope of review, the district court had found enough evidence of causation to get to a jury in a ease where two doctors relied on an “every exposure” theory, but another doctor testified in favor of causation without relying on such a theory, and the plaintiffs’ original opening statement—asserting causation—“further compelled” deference to the trial court’s determination.

*557Maj. Op. at 553. I reject this interpretation, which essentially limits the Certain-Teed holding to the very specific facts of that case. However, even accepting the majority’s narrow read of CertainTeed, I find no reason to come to a different conclusion here. Stallings did not rely solely on the “every exposure” theory. In his deposition, Stallings stated that he had personal knowledge of using Georgia-Pacific products when working as a drywall finisher at two separate companies, R. 180-2 (Stallings Dep. at 25; 39-42) (Page ID #5799; 5803-04), and when working on a two-week project in his own home, id. at 47-50 (Page ID #5805-06).' Anne Ksion-zyk, a corporate representative for Georgia-Pacific, testified that the specific brands that Stallings used contained asbestos. R. 194-3 (Ksionzyk Dep. at 85-90) (Page ID #6584-85). In addition to his testimony regarding the “every exposure” theory, Dr. Strauchen specifically said that Stallings’s exposure to Georgia-Pacific products was a contributory factor to his illness. R. 172-6 (Strauchen Dep. at 38-40) (Page ID #5708). This evidence, when paired with medical testimony regarding the “every exposure” theory, is sufficient to establish legal causation under Kentucky law.

The Kentucky Supreme Court’s adoption of the “every exposure” theory is binding on this court. Stallings made a sufficient showing of legal causation by offering medical testimony that “every exposure” was sufficient to cause his injuries and additional testimony demonstrating exposure to asbestos contained in Georgia-Pacific’s products. Because I would reverse the district court’s grant of summary judgment to Georgia-Pacific, I respectfully dissent.