dissenting.
I dissent from the court’s conclusion that Norcon rebutted the presumption of com-pensability. Norcon’s medical evidence was inconclusive as to whether Siebert’s death was work related. Therefore, I would affirm the Board’s determination in this case.
Substantial evidence rebutting the presumption of compensability consists of “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Thornton v. Alaska Workmen’s Compensation Bd., 411 P.2d 209, 210 (Alaska 1966), quoted in Grainger v. Alaska Workers’ Compensation Bd., 805 P.2d 976, 977 n. 1 (Alaska 1991). Generally, the presumption of com-pensability is not overcome when there is a profound lack of information regarding the cause of an employee’s death. See Fireman’s Fund Am. Ins. Cos. v. Gomes, 544 P.2d 1013, 1017 (Alaska 1976) (unexplained shooting); Employers Commercial Union Co. v. Libor, 536 P.2d 129, 132 (Alaska 1975) (herniated disc).
Libor’s facts are especially pertinent to the case at bar. The employee received treatment for a herniated disc a few weeks after a falling rock struck him in the lower back at his worksite. 536 P.2d at 129. He sought workers’ compensation, and the employer controverted his claim. Id. at 130. The two medical experts who testified were unable to state with reasonable certainty that the back injury was work related, but they did not exclude a causal connection. Id. at 132. One of the doctors stated that he could not “make a definite statement” linking the on-the-job injury with the subsequent herniated disc, although they “could have” been connected. Id. at 130. He went on to state, “I see no reason why there could not be a relationship between the two injuries but it would be almost impossible for me to make the causal connection.” Id. The other doctor stated that the on-the-job injury “cannot, with a reasonable degree of medical certainty, be said to be either the cause of or to have precipitated a herniated disc.” Id. We noted, however, that the other doctor “did not positively exclude a causal connection; he was merely unable to establish it.” Id. at 132. Accordingly, we held that the combination of the employee’s lay testimony and the inconclusive medical testimony constituted substantial evidence that the injury was work related, and that the Board could have concluded that the employer had not overcome the presumption of compensability. See id.
Instead of looking to the Libor rule, here the court relies on two recent decisions, Childs v. Copper Valley Electric Ass’n, 860 P.2d 1184 (Alaska 1993), and Big K Grocery v. Gibson, 836 P.2d 941 (Alaska 1992). The court culls the correct legal rule from these cases: “An employer has always been able to rebut the presumption of compensability with an expert opinion that ‘the claimant’s work was probably not a substantial cause of the disability.’ ” Childs, 860 P.2d at 1189 (quoting Gibson, 836 P.2d at 942). Nonetheless, I believe the court is mistaken in concluding that Norcon met this burden.
Childs and Gibson were cases in which an employer’s medical experts ruled out a link between an employee’s job and his or her injury. Practically speaking, such testimony need not be expressed in absolute or unqualified terms. When experts examine the evidence and become fairly certain about causation, we do not construe their testimony as *1058inconclusive. Instead, we attribute the lack of absolute certainty to the tendency of the scientific mind to avoid unqualified statements on causation. See Childs, 860 P.2d at 1189; Gibson, 836 P.2d at 942; 3 Arthur Larson, The Law of Workmen’s Compensation § 80.32, at 15-834 through -835 (1993).
The opinions of Drs. Scheidt and Allan more closely resemble the medical testimony in Libor than the testimony in Childs or Gibson. Although Dr. Scheidt expressed conjectural doubt that Siebert’s death was work related, he also stated that there is a great deal of uncertainty regarding the causes of heart failure. He also discussed a study that found correlations between lifestyle changes, including working hours, and sudden cardiac death. In regard to the literature on the connection between stress and ventricular arrhythmia, he indicated that there were some studies showing a connection, and some finding none:
There are some studies that show [psychological stress to be a trigger for ventricular arrhythmia]. But exactly in whom and exactly — and again, you know, every study sort of has its own triggers. One has long working hours and another has depression and another has something else. It’s a very confusing literature that doesn’t permit you to come to any definite conclusion. But, your question, are there some studies that show a relationship, yes.
The conclusion of Dr. Scheidt’s deposition testimony is similarly ambiguous:
Most eases that are very common that occur in men just like Mr. Siebert, they occurred for seemingly random and inexplicable reasons. And since so many other cases occur for no reason, since the medical literature in our experience gives us no particular reason, I have to conclude that also in Mr. Siebert’s case ... there is no clear reason and no reason to believe that it is related to his job stress.
(Emphasis added).
Dr. Allan’s testimony is no more conclusive. When asked whether exhaustion was a significant factor in Siebert’s death, Dr. Allan stated that “[t]here really is not the database to answer that kind of question.”
The majority acknowledges that Norcon’s burden was to provide substantial evidence that “directly eliminates any reasonable possibility that employment was a factor in causing the disability.” Inconclusive medical testimony emphasizing a lack of data does not satisfy that burden. Therefore, I would affirm the determination of the Board.