Claimant1 seeks review of a final order of the Workers’ Compensation Board affirming employer’s denial of his claim. He argues that the Board misinterpreted the last injurious exposure rule. We reverse and remand for reconsideration.
At the time of the hearing in 1992, claimant was a 43-year-old auto mechanic suffering from mesothelioma, a cancer of the chest cavity. The only known cause of meso-thelioma is asbestos. Claimant had served in the Navy from 1963 to 1968, where he was exposed to asbestos. After his discharge, he worked as an auto mechanic for 24 years but did not begin working in Oregon until 1984. There is evidence that claimant was exposed to asbestos as an auto mechanic. In 1990, while working for Spectrum Motorwerks, Ltd., claimant sought treatment for chest pains, which were later diagnosed as symptoms of his mesothelioma.
Claimant filed a claim against his Oregon employers, and settled with all of them except Spectrum Motorwerks, Inc., (SMI) and Spectrum Motorwerks, Ltd. (SML).2 Both employers denied the claim and claimant requested a hearing. Dr. Dobrow, claimant’s treating physician and the only medical witness to testify regarding causation, testified that meso-thelioma has a minimum latency period of 10 years. The Board found that the asbestos exposure that caused the mesothelioma must have occurred before 1980. That led the Board to conclude that it was impossible for any Oregon *70employment to have caused claimant’s mesothelioma and the Board affirmed employer’s denial. We review for errors of law, ORS 656.298(6); ORS 183.482(8)(a), and reverse.
In this case, both responsibility and compensability are at issue. In order to establish that his mesothelioma is an occupational disease, claimant must show that it arose “out of and in the course of employment” and was “caused by substances or activities to which [he was] not ordinarily subjected or exposed other than during a period of regular actual employment * * *." ORS 656.802(1). In other words, he must show that work was the major contributing cause of his disease. Runft v. SAIF, 303 Or 493, 498, 739 P2d 12 (1987); Bennett v. Liberty Northwest Ins. Corp., 128 Or App 71, 74, 875 P2d 1176 (1994). In determining whether a claimant’s occupational disease is work-related, the Board should consider all employment exposure, including out-of-state employment exposure. Silveira v. Larch Enterprises, 133 Or App 297, 303, 891 P2d 697 (1995). Here, the evidence as to causation indicates that claimant’s lifetime work-related exposure to asbestos caused his mesothelioma.
The employers contend that, because of the latency period of mesothelioma, they could not actually have caused claimant’s condition and the claim is not compensable as to them. When a claimant has been exposed to the disease-causing substance at more than one employment, questions naturally arise as to the nature and degree of exposure at each employment and how they might bear on the compensability of the claim. The Supreme Court explained the problem in Runft, 303 Or at 499:
“[O]ccupational diseases * * * often develop several decades after first exposure to the disease-causing substance. Although it may be relatively simple to identify the employments at which the claimant was or could have been exposed, it will often be difficult or impossible to establish the nature and degree of the exposure. Similarly, although the relationship between a disease and exposure to a substance may be well established, it will often be difficult to state with sufficient medical probability the degree to which, if any, a particular exposure contributed to the development of the disease.”
*71To alleviate that problem, the courts have adopted the last injurious exposure rule. Under that rule, if a claimant
“proves that a disease was triggered at one time, claimant has carried his burden of proof by establishing that the employer on the risk at the time disability occurred could have caused it, even though previous employers provided conditions which could have caused it, and the rule relieves the claimant of any burden of proving actual causation.” Bracke v. Baza’r, 293 Or 239, 248-49, 646 P2d 1330 (1982).
In Runft, the Supreme Court said:
“The claimant is required to prove only that the disease was caused by employment-related exposure; the claimant is not required to prove that exposure at a particular employer’s workplace caused the disease. Whether employment at any one workplace was the actual cause of the disease is irrelevant under the rule.” 303 Or at 500. (Emphasis supplied.)
Thus, it is immaterial for purposes of establishing the compensability of the claim, that the employers here, because of the latency period of mesothelioma, were not the actual cause of claimant’s disease. All claimant must show to establish a compensable claim is that conditions at the Oregon employer were of the type that could have caused the disease. Accord Bracke, 293 Or at 248-49; Fossum v. SAIF, 293 Or 252, 646 P2d 1337 (1982); Meyer v. SAIF, 71 Or App 371, 692 P2d 656 (1984), rev den 299 Or 203 (1985); but see FMC Corp. v. Liberty Mutual Ins. Co, 70 Or App 370, 689 P2d 1046 (1984), on recon 73 Or App 223, 698 P2d 551, rev den 299 Or 203 (1985).
Employers next contend that, because of the latency period of mesothelioma, conditions of employment at their workplaces could not have contributed to claimant’s disease. Therefore, they argue, and the dissent3 agrees, they cannot be found responsible for claimant’s occupational disease. The *72Supreme Court rejected this very argument in Fossum. Fossum also involved a claimant with mesothelioma. The claimant had worked for three shipbuilders from the early 1940’s to 1948, for W.R. Grasle from 1948 to 1967 and for Willamette Western from 1969 to 1976. In 1977 the claimant died and his widow filed a claim against all the employers. The Board found that the claimant had not proved actual causation against any one employer and held that the claim was not compensable. We reversed, holding that:
“This particular form of cancer does not generally develop until 20 to 40 years after exposure. While it is clear that the deceased was exposed to asbestos at Willamette Western, we are satisfied from the medical evidence that this exposure did not contribute to the cause of his disease in this case and could not have done so.”
Fossum v. SAIF, 52 Or App 769, 774, 629 P2d 857 (1981), aff’d 293 Or 252 (1982). The Supreme Court affirmed this court, “but on different reasoning.” It first noted that conditions at Willamette Western did not involve exposure to asbestos, then stated:
“The Court of Appeals apparently excluded the [Willamette Western] employment because it was not an actual cause. The correct analysis under the last injurious exposure rule, however, is that [Willamette Western] is not liable because there is no evidence of exposure at [Willamette Western] to conditions which could have caused the disease.” 293 Or at 256 (emphasis supplied).
At that point, the Supreme Court included the following footnote:
“The employment at Willamette Western could not have been an actual cause of the disease because it was during the 20-year period before which actual causation must have occurred. Had it been proved and found that the conditions of exposure at Willamette Western were of a kind which could, have caused the disease, then under the last injurious exposure rule as described by Larson, Willamette Western would be liable even though that employment could not have been the actual cause.” 293 Or at 256 n 1 (emphasis supplied).
The present case is almost identical to the hypothetical posed by the court in the footnote in Fossum.
*73We demonstrated our understanding of Fossum’s rejection of employer’s and the dissent’s analysis in Meyer. In Meyer, the claimant sought review of a Board order upholding SAIF’s denial of his claim for asbestosis. SAIF had denied responsibility in Meyer for the same reason that SAIF denies it here: the medical evidence indicated that the exposure did not cause the disease. The doctor in Meyer testified that “[i]n my opinion, that particular exposure would have nothing to do with his current condition.” We stated:
“That the 1978 exposure was not the actual cause of claimant’s present condition does not absolve SAIF from responsibility, for the appropriate inquiry under the last injurious exposure rule is not whether the conditions of the last employment actually caused the disease, but whether those conditions were of a kind which could have caused the disease over some indefinite period of time. Mathis v. SAIF, 10 Or App 139, 499 P2d 1331 (1972); see Fossum v. SAIF, 293 Or 252, 256, n 1, 646 P2d 1337 (1982).”
The dissent attempts to distinguish Meyer by saying “[h]ere, decedent died before any potentially hazardous work conditions with employer could have caused other tumors. ” 135 Or App at 83. The dissent is correct, but that does not distinguish Meyer. In Meyer, there was no doubt that the work conditions at the employer did not and could not have caused the claimant’s disease. The issue was whether the conditions, if exposure was continued over some time, could cause the disease. Here, as in Meyer, the testimony clearly indicates that the conditions could cause the diseases from which claimant suffers. In this case, Dr. Dobrow, plaintiffs oncologist, testified as follows:
“Q. I want you to assume, and this is probably something that will never happen, but assume that [claimant] doesn’t die from the tumor that he has now, this meso-thelioma tumor that he has now.
“You’re not telling us that the exposure that he’s had in April, May, June and July of 1989 could not cause a tumor in the future?
‘ ‘A. No, I’m not saying that — that — that they could not cause a future tumor.
ÍÍ* * * * *
“Q. Assuming that there was exposure then those, in your opinion, are potentially harmful exposures to [claimant]?
*74“A. After a sufficient incubation period, yes.
" * * * * *
“Q. You can tell us that potentially this exposure, if it’s — if it occurred between April and June or July of 1989, was at least potentially causative, potentially harmful to [claimant]?
‘ ‘A. Potentially at some future date, but not dealing with this present tumor.” (Emphasis supplied.)
Dr. Dobrow’s testimony clearly indicates his belief that, given “a sufficient incubation period,” or, as the dissent might put it, “an indefinite period of time,” claimant’s exposure could, in fact, have caused his disease.
If a claimant proves that there was Oregon employment with “conditions of exposure * * * of a kind which could have caused the disease,” then the last employer with those conditions will be assigned responsibility under the last injurious exposure rule. Neither the Board nor the referee reached that question in this case, because they mistakenly determined that, because of the latency period, neither SML nor SMI could have been the actual cause of claimant’s disease.
We recognize that our application of the last injurious exposure rule in this fashion leads to an arbitrary and counter-intuitive result. Nevertheless, since the rule’s inception, the courts, for appropriate policy reasons, have accepted its arbitrary nature. See Inkley v. Forest Fiber Products Co., 288 Or 337, 345, 605 P2d 1175 (1980); Bracke, 293 Or at 248; 4 Larson, Worker’s Compensation Law 17-173, § 95.24 (1994). As the Supreme Court has said:
“[T]he rule can be extremely arbitrary in a particular case, but this arbitrariness is mitigated in the long run as responsibility is spread proportionately among employers by operation of the law of averages.” Runft, 303 Or at 500.
The reason for the rule lies not in the achievement of individualized justice, but rather in its utility in spreading liability fairly among employers by the law of averages and in reducing litigation. Bracke, 293 Or at 248.
The legislature generally has accepted the arbitrary nature of the last injurious exposure rule, except where it has *75decided to change it. As the dissent correctly notes, the legislature recently amended ORS 656.308(1) to read as follows:
“When a worker sustains a compensable injury, the responsible employer shall remain responsible for future compensable medical services and disability relating to the compensable condition unless the worker sustains a new compensable injury involving the same condition. If a new compensable injury occurs, all further compensable medical services and disability involving the same condition shall be processed as a new injury claim by the subsequent employer.” ORS 656.308(1).
The new language “makes it more difficult to transfer responsibility for a condition or disability to a subsequent employer.” SAIF v. Drews, 318 Or 1, 860 P2d 254 (1993). The effect of that amendment was to overturn some aspects of the last injurious exposure rule as developed in Runft, Boise Cascade Corp. and Bracke. Drews, 318 Or at 7 n 3. However, the amendment had no effect on the last injurious exposure rule in an initial claim context. Bennett, 128 Or App at 75 n 1. The 1991 amendments modified one particular application of the last injurious exposure rule, but left most of the rule intact. The dissent does not claim that the amendments changed the last injurious exposure rule enunciated in Meyer and Fossum. Instead, it attempts to assess the legislature’s general intent. 135 Or App at 84. A more plausible reading of the effect of the 1991 amendments is that the legislature had the opportunity to amend the operation of the rule in an initial claims context, but chose not to. Cf. Fifth Avenue Corp. v. Washington Co., 282 Or 591, 597-98, 581 P2d 50 (1978) (amendatory acts do not change the meaning of preexisting language further than is expressly declared or necessarily implied).
The basic effect of the rule is to create a risk pool of all employers whose workplaces have conditions that could cause the disease. As the Supreme Court explained in Bracke:
“[The last injurious exposure rule] is fair to employers only if it is applied consistently so that liability is spread proportionately among employers by operation of the law of averages. * * * [Employers may require consistent application of the rule] so as to assure that they are not assigned disproportionate shares of liability relative to other *76employers who provide working conditions which generate similar risk.” 293 Or at 249-50.
The dissent demonstrates a fundamental misunderstanding of the last injurious exposure rule when it speaks of the rule creating “a risk pool of those employers of the claimant who could have actually contributed to the cause of the claimant’s disease.” 135 Or App at 86. To speak of “the law of averages” spreading liability proportionately only “among the employers who exposed the claimant to similar risks” makes no sense. Under the last injurious exposure rule, only one of a particular claimant’s employers is found liable and all other employers have no liability. Under the dissent’s concept of the risk-pool, there is no “spreading of liability.” The dissent apparently recognizes the weakness of its artificial “risk pool” construct and posits an alternative reason why the correct conception of the risk pool is still unfair. 135 Or App at 86 n 2. However, that argument is the same one that Judge Deits makes in her dissent here and in Silueira. We see no reason to revisit our earlier decision.
The dissent later castigates the majority for “empowering] the imposition of responsibility on an innocent employer” and says that the net result makes
“an Oregon employer responsible for a disease that was like a time bomb in decedent before he ever came to work in Oregon and to which the Oregon employer did not contribute or cause. That’s not fair!”
The dissent might have a point, if employers held liable under the last injurious exposure rule were truly innocent. However, the employers subject to liability under the rule are only those with harmful working conditions that could cause the disease. Liability attaches to an Oregon employer only if it is creating its own time bombs.
Finally, the dissent argues that:
“There is nothing in the Workers’ Compensation Act to suggest that the legislature ever envisioned that an Oregon employer could be responsible for out-of-state employment conditions based on the law of averages among employers world-wide.” 135 Or App at 87.
*77The dissent is correct, but we note that there is also nothing in the Workers’ Compensation Act to suggest that the legislature ever envisioned that an Oregon worker with an occupational disease caused entirely by his working conditions could go without compensation. In the light of the long-standing policy that the Workers’ Compensation Act is to be liberally construed for the benefit of the worker, Fossum v. SAIF, 289 Or 787, 805, 619 P2d 233 (1980); Liberty Northwest Ins. Corp. v. Short, 102 Or App 495, 499, 795 P2d 118 (1990); Holden v. Willamette Industries, 28 Or App 613, 618, 560 P2d 298 (1977), and the fact that the legislature has not specifically addressed this question, we fail to understand the dissent’s strained attempts to ignore controlling precedent.4
The last injurious exposure rule requires that this case be remanded to the Board for consideration of whether SMI or SML are responsible for the claim because conditions at those employments were of a kind that could have caused the disease.
Reversed and remanded for reconsideration.
William A. Strametz originally filed the claim at issue in this case. He pursued his claim through the hearing before the referee and before the Board on review. William Strametz has since died. The beneficiaries of his estate are allowed to proceed with his claim under ORS 656.218(3), which provides:
“If the worker has filed arequest for a hearing pursuant to ORS 656.283 and death occurs prior to the final disposition of the request, the persons described in subsection (5) of this section shall be entitled to pursue the matter to final determination of all issues presented by the request for hearing.”
Thus, although we use the term claimant in referring to Strametz, itisforeaseof reference. The actual claimants are the beneficiaries of his estate.
SMI was an Oregon business formed by claimant and his wife in the mid 1980’s. In July 1988, SMI became insured by SAIF, but the insurance did not provide coverage to claimant because of his status as a partner in SMI. See ORS 656.027(8). In July 1989, claimant and his wife sold the business and it was renamed SML. Claimant continued to work at SML as a subject employee. SML was later declared to be a noncomplying employer.
There are two dissents to this opinion. Judge Deits dissents for the same reasons she dissented in Silveira v. Larch Enterprises, 133 Or App 297, 891 P2d 697 (1995). Judge Edmonds dissents for fundamentally different reasons that were not at issue in Silvera. Judge Deits’ concerns were answered in the majority’s opinion in Silveira. Thus, when we refer to the dissent in this opinion, we are referring solely to Judge Edmonds’ dissent.
We note that we are not the first state to come to this conclusion and also note what the Nebraska Supreme Court has said:
“[E]ven though liability imposed under this rule can have a harsh result [in a particular case], there will be a spreading of the risk when the total picture of asbestos litigation is considered on a nationwide basis.” Osteen v. A. C. & S., Inc., 209 Neb 282, 307 NW 2d 514 (1981).