OPINION
BRYNER, Justice.
Daretha Tolbert filed four workers’ compensation claims against her employer, Alas-com, Inc., alleging work-related injuries to her hands. The Alaska Workers’ Compensation Board denied these claims, finding that Tolbert either had shown no compensable injury or had failed to prove that her injuries were work related. The superior court affirmed. As to three of the claims, we conclude that Tolbert presented sufficient evidence to raise a presumption of work-related injuries and that Alascom failed to rebut this presumption.
I.
FACTS AND PROCEEDINGS
A.
Tolbert’s Original Claim
Daretha Tolbert began working for Alas-com in 1982 as a telephone operator. Her job required at least some repetitious use of a keyboard, although the parties dispute how much keyboarding was involved. Between 1982 and 1989, she developed bilateral carpal tunnel syndrome, requiring her to undergo multiple carpal tunnel release surgeries. Tolbert sought to recover workers’ compensation for what she maintained was a job-related impairment.
The Board found that her claims were barred by AS 23.30.100(a) and (d) as a result of her delay in notifying Alascom that her injuries were job related. The superior court affirmed the Board’s decision in July 1991; we affirmed the superior court’s holding in March 1992.
B.
Tolbert’s Subsequent Claims in
1992-
1991
Although our March 1992 decision affirming the denial of Tolbert’s original claim barred her from any potential recovery based on her existing carpal tunnel syndrome, it did not bar her from claiming benefits for subsequent work-related incidents that aggravated her condition or caused new injuries to her hand. Between 1992 and 1994, Tolbert filed four claims with the Board, alleging that her continued work with Alas-com aggravated her existing hand condition or otherwise hurt her hands.
In one claim, Tolbert reported a mishap at work in June 1992 in which an opening door slammed against her right hand. In three other claims, she asserted that repetitive hand movements required by her keyboarding job caused her to experience pain, numbness, and swelling in her hands in February of 1993 and on June 1 and 14,1994. In these latter claims, Tolbert alleged permanent impairment in the form of tendinitis or aggravation of her carpal tunnel syndrome; she sought reimbursement for medical care related to these problems and benefits for temporary total and permanent partial disability. Following a hearing, the Board denied all of her claims. The superior court affirmed. Tolbert appeals.
II.
DISCUSSION
A.
Standards of Review
When the superior court has acted as an intermediate court of appeal, we review the merits of the administrative agency’s decision without deference to the superior
court’s decision.
We review questions of law involving agency expertise under the “reasonable basis” test
and those not involving any particular • expertise under the substitution of judgment standard.
And we review determinations of fact by an administrative agency under the “substantial evidence” standard,
but consider “[t]he question of whether the quantum of evidence is substantial [to be] a legal question.”
B.
The Board Did Not Err in Denying TolheH’s Claim for the 1992 Injury.
Tolbert alleged that her right hand was injured in June 1992 by an opening door. She sought compensation for medical payments and an award of benefits for permanent partial disability. The Board held that Tolbert had raised a presumption of com-pensability as to this claim by showing that she had suffered either a contusion or a possible aggravation of tendinitis; the Board further found that Alascom had failed to overcome this presumption. The Board nevertheless declined to award Tolbert any benefits, concluding that she had failed to produce sufficient evidence to support an award.
1.
The Board did not err in denying medical benefits to Tolbert.
The only evidence Tolbert presented to prove medical expenses stemming from her June 1992 injury consisted of her own testimony; she offered little more than a recounting of the money she owed to certain medical providers. But Alaseom’s claims adjuster denied having ever received any of Tolbert’s medical bills. While the Board viewed the adjuster’s testimony as credible, it found Tolbert’s testimony both vague and lacking in documentary support. The Board further determined that, because Tolbert failed to submit the bills to Alascom at all, much less in the form required by 8 Alaska Administrative Code (AAC) 45.082(d), and, independently, because she failed to submit such bills to the Board, it lacked sufficient evidence upon which to base an award of medical benefits.
Tolbert argues that she did not fail to comply with 8 AAC 45.082(d), because that subsection only governs the form and not the time of submission of medical bills to the employer. According to Tolbert, the Board should have left the hearing record open to allow Tolbert to submit her medical bills in proper form. Alascom replies that Tolbert has waived the right to submit these bills by failing to request either a continuance of the hearing or that the record remain open so that medical bills could be submitted. We conclude that the record supports the Board’s ruling that Tolbert failed to present sufficient evidence on this issue.
A presumption of compensability applies to all workers’ compensation claims;
if the presumption remains unrebutted, the Board must find that the claim is compensa-ble.
The worker is thus freed from having to prove (1) that “but for” the employment the disability would not have occurred, and (2) that reasonable persons would regard the employment as a cause of the injury and attach responsibility to it.
But the presumption of compensability does not free an injured worker from the burden of introducing evidence as to the extent of the injury and the amount of medical expenses.
Allocation of this burden to the claimant makes sense because the extent of injury and amount of medical expenses are unique in each case, and the worker often has greatest
access to such information.
Because medical expenses are not presumed, a claimant has the burden of proving them by a preponderance of the evidence.
Tolbert failed to meet this burden. Although she testified that she owed certain medical professionals money, she failed to link these debts to treatment of the 1992 injury. From the time of injury in June 1992 to her hearing in June 1995, she had ample time to gather this evidence. Despite this three-year interval, she obtained no documentation of her medical expenses. At the time of the heating, Tolbert’s counsel expressed belief that the bills either had already been placed in the record or were in Alascom’s possession. But even after discovering otherwise, Tolbert’s counsel failed to submit the bills or even to request that the record remain open to accommodate later submission. Instead, counsel opted to proceed with Tolbert’s testimony as the only evidence on expenses.
We conclude that, given these circumstances, the Board did not abuse its discretion by failing to order sua sponte that the record remain open for Tolbert to submit her medical bills.
2.
The Board did not err in denying Tolbert’s claim, of permanent partial impairment.
Tolbert argues that the Board’s refusal to award permanent partial impairment (PPI) benefits for the June 1992 incident is not supported by the record. But the experts who testified about the incident agreed that its effects were only temporary and did not aggravate Tolbert’s carpal tunnel syndrome. Tolbert’s own physician, Dr. Lipke, testified that the 1992 injury caused a contusion. He also testified that the contusion could lead to tendinitis, but he never attributed Tolbert’s tendinitis to this incident, and he described her tendinitis as intermittent and recurring, rather than permanent. In fact, Dr. Lipke expressly testified that tendinitis translates into no PPI rating under the
American Medical Association Guides to the Evaluation of Permanent
Improvement.
Yet such a rating is a prerequisite to an award of PPI benefits.
Given Dr. Lipke’s testimony, we conclude that the Board’s refusal to award Tolbert PPI benefits is supported by substantial evidence.
C.
The Board Erred in Denying Tolbert’s 199S and 199U Claims.
1.
The testimony concerning Tolbert’s carpal tunnel syndrome and tendinitis
Tolbert filed one claim in February 1993 and two claims in June 1994, asserting that the repetitious keyboarding she performed for Alaseom was causing swelling, numbness, and pain in her hands and arms. Tolbert initially asserted that these symptoms were caused by aggravation of her pre-existing carpal tunnel syndrome.
Ultimately, however, she was able to present little evidence to support this theory. But she did present considerable evidence indicating that the problems she experienced stemmed from work-related episodes of tendinitis.
At the hearing on her claims, Tolbert described suffering from work-related problems with her hands on February 22, 1993, June 1, 1994, and June 14, 1994. She presented extensive expert testimony suggesting that the problems she experienced were work related. Dr. Lipke, Tolbert’s treating physician and primary witness, testified about carpal tunnel syndrome and tendinitis, describing the differences between the two conditions. According to Dr. Lipke, carpal tunnel is the narrow passageway in the wrist between the wrist bones and the transverse carpal ligament. The flexor tendons and the median nerve must pass through this tunnel to reach the fingers. Although the causes of carpal tunnel syndrome vary, one cause is repetitious trauma. Repetitious motion, coupled with the normal aging process, can cause pain in, and swelling of, the tendons — • resulting in tendinitis. Tendinitis can, by putting pressure on the median nerve, in turn, lead to carpal tunnel syndrome. Tendinitis is a temporary condition and generally will resolve without causing carpal tunnel syndrome. But if the pressure is persistent enough, nerve dysfunction, pain, numbness, and tingling in the fingers — collectively called carpal tunnel syndrome — will result. Scar tissue may eventually form on the nerve, which can, when combined with external pressure, continue to irritate the nerve even after the tendinitis resolves. Once the carpal tunnel syndrome becomes sufficiently severe, surgery is necessary to release it. Carpal tunnel release surgery consists of cutting the ligament that forms part of the carpal tunnel, thereby releasing the pressure on the median nerve.
Dr. Lipke also addressed the issue of Tol-bert’s diagnosis. He commented that, in Tolbert’s case, nerve conduction studies revealed no aggravation of her carpal tunnel syndrome. In Dr. Lipke’s view, Tolbert’s problems were more likely attributable to tendinitis. Specifically, Dr. Lipke testified that repetitive motions like keyboarding can cause and aggravate tendinitis, that Tolbert’s symptoms were consistent with tendinitis, and that tendinitis could be objectively confirmed if volumetric tests showed abnormal swelling.
Volumetric testing of Tolbert’s hands did in fact reveal abnormal swelling. Dr. Ferris conducted two volumetric tests on Tolbert. The first simulated Tolbert’s job as she had described it, but Dr. Ferris later deemed this test unreliable because he learned that Tol-bert had exaggerated the amount of keyboarding she usually performed. Dr. Ferris personally designed the second test after speaking with Alascom about the kinds of work Tolbert actually did and after observing Tolbert’s work being performed by other Alascom workers. Dr. Ferris testified that this second test accurately simulated Tol-bert’s job conditions. In both tests, Tolbert’s hands swelled abnormally.
Dr. Ferris, like Dr. Lipke, found the results of the volumetric testing significant because the tests had been designed to mimic Tolbert’s actual work conditions and they produced abnormal swelling. In Dr. Ferris’s view, the objective signs of swelling substantiated Tolbert’s subjective complaints of hand pain. Moreover, swelling to the degree he observed would cause pain and “decrease the amount of function that [Tolbert] would feel comfortable with carrying out.” He indicated that he believed that Tolbert’s hands would continue to swell as long as she persisted in the sorts of activities required by her job.
Although Dr. Ferris said that Tolbert’s hand swelling was not necessarily tendinitis and might be a sort of passive edema related to carpal tunnel syndrome, he recognized that his opinion on this issue was speculative, and he deferred to Dr. Lipke’s diagnosis of tendinitis, because, in Dr. Ferris’s view, Dr. Lipke had more expertise in diagnosing the condition.
Two other experts presented evidence at Tolbert’s hearing: Drs. Sack and Fu. Their testimony aimed primarily at negating the possibility that Tolbert’s problems had been caused by an aggravation of her carpal tunnel syndrome. Both agreed that her work had not caused any substantial permanent aggravation in her carpal tunnel syndrome.
Dr. Fu testified that Tolbert was left with mild residual nerve entrapment after her carpal tunnel release surgeries and, as a result, her “symptoms” might increase with
activity. But he did not explain what symptoms he expected her to have. He acknowledged the presence of some form of “temporary aggravation and reaggravation of pain and discomfort” when Tolbert engages in “repetitive hand activities.” But beyond disputing Tolbert’s claim of aggravated carpal tunnel syndrome and mentioning the possibility of some residual “symptoms” that might occur in the absence of aggravation, Dr. Fu did not specify what Tolbert’s problem might or might not be. While he did testify that she still objectively showed some “decrease of sensations involving both hands” and increased sensitivity along her wrist, he drew no inferences from these symptoms. As for the swelling in Tolbert’s wrists, he failed to account for it at all.
Like Dr. Fu, Dr. Sack also unequivocally testified that he did not believe that Tolbert’s carpal tunnel syndrome had been aggravated. But he, too, was silent with respect to tendinitis and did not account for Tolbert’s swelling. He testified that Tolbert’s carpal tunnel syndrome, even after surgery, would cause a “bit of symptomatology.” But he did not describe what symptoms he had in mind; nor did he deny that Tolbert might also be suffering from tendinitis.
2.
Tolbert established the preliminary link necessary to raise the presumption that her employment substantially aggravated her tendinitis.
The Board found that there was “no medical testimony indicating [that Tolbert] suffered a work-related injury on February 22, 1993, June 1, 1994 or June 14, 1994.” Accordingly, it concluded that Tolbert failed to establish the preliminary link necessary to trigger the presumption of compensability. Tolbert challenges this finding.
Alaska Statute 23.30.120 presumes that workers’ compensation claims are compensable. But this presumption does not apply automatically; we have held that the worker must show a preliminary link between the injury and the job.
To establish such link, “the claimant need not present substantial evidence that his or her employment was a substantial cause of ... disability.”
Rather, an offer of “some evidence” that the claim arose out of the worker’s employment is sufficient.
For purposes of determining whether the claimant has established the ' preliminary link, only evidence that tends to establish the link is considered — competing evidence is disregarded.
Likewise, credibility plays no part in the process: “In making its preliminary link determination, the board need not concern itself with the witnesses’ credibility.”
At her hearing, Tolbert reported work-related problems with her hands occurring on February 22, 1993; June 1, 1994; and June 14,1994. She presented a great deal of expert testimony suggesting that the problems she experienced were work related. Dr. Lipke testified that keyboarding can cause and aggravate tendinitis, that Tolbert’s symptoms were consistent with tendinitis, and that tendinitis could be objectively confirmed if volumetric tests showed abnormal swelling — which, in Tolbert’s case, they did.
Considering the overwhelming medical evidence supporting a work-related injury, and given the minimal showing required to establish the preliminary link and the irrelevance of credibility at this phase of the inquiry, we hold that there was sufficient evidence to establish the preliminary link and give rise to the presumption of compensability. Thus, the Board erred in finding that Tolbert did not produce “some evidence” linking her injuries to her job.
3.
Alascom did not present substantial evidence to overcome the presumption that Tolbert’s tendinitis was work related.
Once the preliminary link has been established, “it is the employer’s burden to overcome the presumption [of compensa-bility] by coming forward with substantial evidence that the injury was not work related.”
Substantial evidence is evidence that a reasonable mind, viewing the record as a whole, might accept as adequate to support the Board’s decision.
We have explained this standard as follows:
To overcome the AS 23.30.120(a) presumption of compensability, an employer must present substantial evidence that either “(1) provides an alternative explanation which, if accepted, would
exclude
work-related factors as a substantial cause of the disability; or (2) directly eliminates
any reasonable possibility
that employment was a factor in causing the disability.”[
]
If the employer successfully rebuts the presumption of compensability, the presumption drops out, and the employee must prove all of the elements of the ease by a preponderance of the evidence.
Here, the Board found that, even if Tolbert had established the preliminary link, Alas-com would have overcome “the presumption of compensability with the testimony of Dr. Sack and Dr. Fu.” But, again, the thrust of both doctors’ testimony was that Tolbert had experienced no permanent aggravation of her carpal tunnel syndrome — they scarcely referred to the issue of tendinitis. And while the doctors mentioned the possibility that Tolbert’s prior carpal tunnel syndrome might continue to manifest itself in a “bit of symp-tomatology,” neither specified what symptoms might occur or characterized such symptomatology as a likely cause of Tolbert’s recent problems. In short, although Dr. Sack and Dr. Fu did point to a possible alternative cause for Tolbert’s pain, they did not describe this alternative as its probable cause or otherwise attempt to rule out Dr. Lipke’s diagnosis of work-related tendinitis.
We have previously recognized that, for purposes of overcoming the presumption of compensability, “medical testimony cannot constitute substantial evidence if it simply points to other possible causes of an employee’s injury or disability, without ruling out work-related causes.”
Because Dr. Sack’s and Dr. Fu’s testimony does nothing more than point to other possible causes, it fails to rebut the presumption of compensability.
4.
The Board erred in alternatively concluding that Tolbert had not proved that her claims were work related because she failed to show “but for” causation.
The Board alternatively concluded that Tolbert failed to prove her claims by a preponderance of the evidence because she failed to prove that her wrist injuries would not have occurred “but for” the work. Though the Board accepted Dr. Lipke’s testimony that the swelling in Tolbert’s hands indicated a problem with tendinitis and that Tolbert’s work at Alascom aggravated this condition, it nevertheless found that her activities at home also might have aggravated the condition. Therefore, the Board reasoned, Tolbert had failed to establish that, “but for” her work, she would not have been disabled.
But this reasoning conflicts with our
prior
holdings on substantial-factor causation. We have said that, “when two or more forces operate to bring about an injury and each of them, operating alone, would be sufficient to cause the harm, the ‘but for’ test is inapplicable because it would tend to absolve
all forces from liability.”
In such cases, it is necessary to ask whether the work-related injury was a substantial factor in causing the disability: “If one or more possible causes of a disability are [work related], benefits will be awarded where the record establishes that the [work-related] injury is a substantial factor in the employee’s disability regardless of whether a [non-work-related] injury could
independently
have caused disability.”
The “but for” standard applied by the Board required Tolbert to prove that her work-related injury was the sole cause of— and not merely a substantial factor in causing — her disability. Because the standard applied by the Board conflicts with the substantial-factor test, which we have held applicable in cases like Tolbert’s, we reject the Board’s alternative ground for concluding that Tolbert failed to prove her claims.
5.
The Board erred in alternatively concluding that Tolbert’s claims are barred by res judicata.
As yet another independent ground for denying Tolbert relief, the Board concluded that her 1993 and 1994 claims were not compensable because they were legally precluded. Finding that Tolbert’s tendinitis was as longstanding as the underlying carpal tunnel syndrome, the Board reasoned that any claims relating to her tendinitis were barred by res judicata following our 1992 decision affirming the denial of Tolbert’s original claims for carpal tunnel syndrome. But this holding conflicts with the Board’s finding that tendinitis is a temporary (albeit recurring) condition. Under these circumstances, claim preclusion does not apply.
III.
CONCLUSION
We AFFIRM the superior court’s judgment as to Tolbert’s June 1992 injury. But as to her 1993 and 1994 claims, given that Tolbert raised the presumption of compensa-bility and that Alascom failed to rebut it, we must REVERSE the Board’s decision and REMAND this case for a determination of the extent and compensability of Tolbert’s injuries.