OPINION
I. INTRODUCTION
John Orchitt was exposed to radio frequency radiation in an accident while he worked for AT & T Alascom. After a contested hearing, the Alaska Workers’ Compensation Board awarded him temporary total disability and medical benefits. AT & T unsuccessfully appealed to the superior court, alleging that procedural irregularities deprived it of due process and that the board’s decision was not supported by competent scientific evidence. Because substantial evidence supports the board’s findings and because the board's procedural decisions did not deprive AT & T of due process, we affirm the superi- or court’s judgment that affirmed the board’s ruling.
II. FACTS AND PROCEEDINGS
John Orchitt applied for workers’ compensation benefits on September 21,1999, claiming he had suffered head, brain, and upper body injuries as a result of overexposure to radio frequency radiation on November 16, 1998. AT & T Alascom controverted his claim on October 14, 1999. We derive the facts in this ease from the workers’ compensation file and hearing record.
Orchitt began working for AT & T Alas-com in 1991, after serving in the Air Force for more than twenty years.1 He worked primarily as a telecommunications equipment installer technician.
On November 16, 1998, Orchitt and his coworker, Tim Sorenson, were installing a new computer-operated switching system in the Eagle River Earth Station. They had to replumb sections of waveguide as part of the installation process.2 To prevent them from being exposed to radio frequency radiation, the amplifier associated with the waveguide they were working on was supposed to be turned off.
After a technician from the Eagle River Earth Station turned off an amplifier in accordance with the specifications provided, Orchitt separated two segments of the waveguide. He estimated that his head was from nine to fifteen inches away from the waveguide’s point of separation. While Orchitt *1236was working on the waveguide, Sorenson walked around the room with a meter and probe to detect any radio frequency radiation. The meter Sorenson used had three scales. A full-scale reading on the highest scale could indicate the presence of three times the American National Standards Institute (ANSI) limit for whole body exposure. Sorenson calibrated the meter outside the room. After he reentered the room, the meter “pegged,” indicating that there was radio frequency radiation in the room. “Pegged” means the meter registered at its highest level. Sorenson changed the scale while he was in the room, but the meter continued to peg. Realizing there was a problem, Orchitt clamped the two pieces of the waveguide together to stop the radiation from leaking any further. Orchitt and Sor-enson then discovered that the amplifier connected to the waveguide had not been turned off because the engineer had misidentified which amplifier was associated with the waveguide they were working on. Orchitt contacted the engineer and tried to contact his supervisor to tell them about the accident; his supervisor was not in, so he contacted the manager instead. The radio frequency radiation Orchitt was exposed to had a frequency of six gigahertz;3 the amplifier transmitting radio frequency radiation through the waveguide was operating at approximately ninety watts. Orchitt estimated that he was exposed to radio frequency radiation for three to six minutes.
Sorenson testified that Orchitt said he felt a “heat flash.” Sorenson did not observe any redness on Orchitt’s face at that time. Or-chitt filed a report of injury on December 14, indicating that his head and eyes had been exposed to radiation. He continued to work as an installer for AT & T for about three months following the accident; some of his work was overtime.
Radio frequency radiation is non-ionizing radiation, unlike the radiation from x-rays. The primary biological effect of radio fre-queney radiation is heating. Ionizing radiation, in contrast, has sufficient energy to break molecular bonds within the body. Radio frequency radiation encompasses a number of frequencies, including the frequencies for television, radio broadcasting, and telecommunications. The term “microwave radiation” refers to a region within the radio frequency radiation band. The frequency of microwave radiation is usually above one gigahertz, or one billion cycles per second. Different frequencies of radio frequency radiation have differing abilities to penetrate tissue. Frequency and wavelength are related, so that longer waves have lower frequencies. Longer waves have greater penetration. Six gigahertz waves penetrate to approximately eight millimeters. When the waves reach this depth, they have lost approximately eighty-five percent of their energy.
Safety standards for exposure to radio frequency radiation vary according to the frequency involved. There are two ways to calculate exposure to radio frequency radiation. One way is to calculate the actual exposure level in milliwatts per square centimeter; the second way measures tissue absorption of radio frequency radiation in watts per kilogram. The Federal Communications Commission (FCC) has set standards both for general public exposure and for occupational exposure. Two experts who testified in Orchitt’s case and the board used the FCC occupational standard for actual exposure to evaluate whether he was overexposed to radio frequency radiation. The FCC occupational standard for actual exposure at six gigahertz is five milliwatts per square centimeter over a six-minute interval for whole body exposure.4
Orehitt’s first medical visit after the exposure was an appointment with his family clinic on November 30, 1998. Orchitt was concerned about headache and eye pain after the exposure but thought he had a sinus infection. The doctor he saw referred Or-*1237chitt to an optometrist for follow up. The optometrist found nothing wrong but referred Orchitt to a neurologist to rule out a stroke. The neurologist ordered an MRI; it showed “tiny areas of hyperintensity in the frontal lobes,” which the neurologist concluded had “doubtful clinical significance.” The neurologist prescribed medication for Or-chitt’s headaches. Dr. David Swanson, an ophthalmologist, evaluated Orchitt’s eyes in February 1999 and found no abnormality except decreased tear production. Orchitt went to Dr. Stanley Smith, his family physician, in March 1999 with complaints about “mental slowing.” Dr. Smith was concerned that Orchitt had suffered a stroke or transient ischemic attack.5
In March 1999 Orchitt’s neurologist referred him to Dr. Marvin Ziskin, a professor of radiology and medical physics at Temple University in Philadelphia. Dr. Ziskin did not examine Orchitt in person, but, using information Orchitt provided him, made calculations related to the amount of radio frequency radiation for Orchitt’s exposure. Based on those calculations, Dr. Ziskin concluded that Orchitt was overexposed to radio frequency radiation.
Dr. Ziskin’s conclusions differed from those of Kimberly Kantner, AT & T’s radiation safety officer.- Following the injury report, Kantner had calculated Orchitt’s probable exposure level, using a mathematical model. Based on these calculations, she estimated a range of radiation exposure levels, with the high end being slightly in excess of the FCC maximum permissible limit. But because of the physical symptoms he described, she concluded that Orchitt had not been overexposed.
Orchitt consulted Dr. Paul Craig, a neu-ropsyehologist, in August 1999. Dr. Craig’s evaluation showed a relatively normal neuro-cognitive profile, although he noted “a very mild neurocognitive disorder” and a “significant level of depression.” Dr. Craig’s report stated that he did not have the necessary expertise to determine whether there was any link between Orchitt’s symptoms and his radio frequency radiation exposure.
Orchitt began treatment at the Brain Injury Association of Alaska in October 1999. His main care provider there was Dr. Debra Russell; she has a Ph.D. in psychology, but is not a licensed clinical psychologist. Dr. Russell conducted some testing on Orchitt and issued an opinion letter to the claims adjuster, stating that Orchitt was suffering from a cognitive disorder as a result of his radio frequency radiation exposure. She provided Orchitt with ongoing rehabilitation therapy to address his continuing complaints of mental slowing and mood changes.
Dr. Russell also referred Orchitt to Dr. Daniel Amen, a psychiatrist, for a single photon emission computed tomography (SPECT) scan. A SPECT scan measures blood flow in the brain, looking at functional, rather than structural, changes.6 Dr. Amen performed the SPECT scan in November 2000 and concluded that Orchitt had some decreased brain activity as well as depression. Dr. Amen attributed the neurological impairments he observed to radio frequency radiation exposure based on the history Or-chitt gave and a discussion Dr. Amen had with Dr. Russell.
AT & T retained a panel of doctors to evaluate Orchitt. Dr. Patricia Sparks, a specialist in occupational and environmental medicine and internal medicine, examined Orchitt in September 2000. Dr. Sparks concluded that while Orchitt may have had some warming of his skin due to the radio frequency radiation exposure, the symptoms he described were not consistent with the known health effects of radio frequency radiation exposure. She believed that Orchitt was suffering from depression that was not directly related to the radio frequency radiation exposure.
Dr. David Coppel, a Washington neuropsy-chologist, also evaluated Orchitt for AT & T in September 2000. He did some of the same testing Dr. Craig had done in 1999. The testing showed some impairments in vi*1238sual processing, but Dr. Coppel did not believe that they could be related to the radio frequency radiation exposure. He instead believed that depression was the most likely cause of Orchitt’s difficulties, but he did not offer an opinion as to the origin of the depression.
Dr. Douglas Robinson, a Seattle psychiatrist, conducted a psychiatric evaluation of Orchitt for AT & T, also in September 2000. He concluded that the late onset of symptoms reported by Orchitt indicated that the radio frequency radiation exposure was an unlikely cause of Orchitt’s difficulties. His opinion stated that the most likely explanation for Orchitt’s complaints was depression and somatization due to stress. He identified several stressors that could have contributed to the depression.
Because of the complex medical issues, the board ordered a second independent medical evaluation (SIME) in April 2000. The board selected Dr. Charles Sutton, a neurosurgeon, to conduct the evaluation. Dr. Sutton spoke with Orchitt by phone and was provided extensive medical records. Dr. Sutton asked the board to hire an engineer as a consultant to give him a better idea of how much radio frequency radiation Orchitt had been exposed to.
At Dr. Sutton’s request, the board hired Dr. Arthur Guy, a professor emeritus of electrical engineering at the University of Washington. Dr. Guy had done extensive work in the area of the biological effects of radio frequency radiation. He conducted three computer models of the accident. The first was based on information received from AT & T. After the report based on the first model concluded that there was no overexposure, Orchitt wrote to Dr. Guy, describing the incident. Dr. Guy then ran a second model, using the information that Orchitt provided. This scenario also showed that Orchitt had not been overexposed to radio frequency radiation. Orchitt again contacted Dr. Guy and supplied other information. Dr. Guy then made a third set of calculations. Because there was conflicting evidence about the placement of possible reflectors, Dr. Guy placed the reflectors in what he considered to be the worst possible placement in terms of radiation exposure. The third scenario produced an exposure that was approximately nine and a half percent over the FCC exposure limits, but not enough to cause biological effects. After receiving Dr. Guy’s reports, Dr. Sutton concluded that Orchitt had not suffered any injury related to the radio frequency radiation exposure beyond dermal heating, which Orchitt experienced as a sensation akin to sunburn.
Orchitt’s board hearing was scheduled to begin on April 8, 2003. The parties attended a pre-hearing conference on March 10, 2003. Orchitt stated at that conference that he would be submitting two new expert reports, one from Dr. Russell, and one from a newly identified expert, Dr. James May, a neurop-sychologist. Dr. May’s report concluded that Orchitt suffered from organic personality syndrome and mood disorder due to general medical conditions and that these conditions related to his exposure to radio frequency radiation. In spite of AT & T’s objection, the board refused to exclude the reports because the board reasoned that they were filed within the twenty-day deadline for filing evidence.7
A short time later, AT & T requested a continuance so that it could get a follow-up employer’s medical examination (EME) of Orchitt in response to the new expert reports. AT & T later withdrew that request with the understanding that Orchitt would attend an EME before the hearing. AT & T scheduled the EME for April 1-3, 2003 in Seattle; however, in a March 19 letter from his attorney, Orchitt indicated that he would not be able to attend the EME as scheduled. AT & T requested a board hearing to address several issues, including the EME and AT & T’s renewed request for a continuance.
At a board hearing on April 1, Orchitt said he would submit that day another new expert report from another new expert, Dr. Jeff Keene, a neuro-ophthalmologist. Dr. Keene diagnosed several vision disorders in Orchitt and made recommendations for treatment. AT & T told the board that it had not yet received all the information it had requested *1239from Dr. May and that Dr. May had not appeared at his scheduled deposition. AT & T asked the board to either strike Dr. May’s report or, alternatively, grant a continuance to allow AT & T to (1) demand that Dr. May release his report and (2) compel Orchitt to attend a follow-up EME. The board denied AT & T’s request to strike Dr. May’s report, decided that it would be unreasonable to require Orchitt to attend a follow-up EME so close to the April 8 main hearing, and reserved ruling on AT & T’s continuance request.
At the April 8 hearing AT & T renewed its request for a continuance for the purpose of developing expert testimony to rebut what it called Dr. May’s “alleged” report. The board denied the request but told AT & T’s attorney that she could renew the request or make objections the following day.
At the time of the hearing Dr. Keene was out of state and unavailable to testify. The board admitted Dr. Keene’s report and said it would hold the record open for rebuttal or cross-examination. Although the board .had admitted Dr. May’s report into evidence, and although Dr. May was available, Orchitt did not present him as a witness. AT & T objected to a board ruling that if AT & T wanted to cross-examine Dr. May, it would have to do so during the time allotted for its ease-in-chief. AT & T’s counsel said that she wanted to think about whether she wanted to take the time from her case-in-chief to cross-examine Dr. May. AT & T never thereafter renewed its request to cross-examine Dr. May.
At the end of the hearing AT & T objected to the denial of its continuance request and also stated that it would waive cross-examination of Drs. May and Keene. The board chair told AT & T that the board would not leave the record open for a follow-up EME and that the board was closing the evidentia-ry record at that time. AT & T did not object to closing the record.
The board’s post-hearing decision and order found that Orchitt had been exposed to excessive amounts of radio frequency radiation. It found that the models of Dr. Guy and Kimberly Kantner did not correspond with the “known facts” in the case. In finding that Orchitt had been overexposed, it relied on the testimony of Orchitt, Sorenson, and Dr. Ziskin. The board decided that Orchitt’s mental deficits and depression were the result of the overexposure. Besides testimony from medical experts, the board relied on testimony from Orchitt’s coworkers that Orchitt had a red face following the accident and that his mental and cognitive states changed after the accident. It also decided that Orchitt’s predominant cause of disability was his depression and awarded him temporary total disability (TTD) benefits through April 21, 2001, the date on which Orchitt applied for unemployment 'benefits and certified that he was available for work. One member of the panel dissented, concluding that Orchitt’s exposure caused only dermal symptoms that readily healed and that AT & T had paid all benefits due Orchitt.
AT & T appealed to the superior court, contending that the board violated AT & T’s due process rights and that the decision was not supported by substantial evidence. AT & T alleged for the first time in the superior court that the board chair was biased. The superior court affirmed the board’s decision, finding that the decision was supported by substantial evidence and that AT & T’s due process rights had not been violated. AT & T appeals.
III. DISCUSSION
A. Standard of Review
We directly review the board’s ruling.8 Whether the board denied AT & T due process is a question of law that does not involve agency expertise; we substitute our judgment and exercise independent review of questions of law.9 We review the board’s factual findings under the substantial evidence standard.10 Substantial evidence is *1240such evidence as a reasonable mind might accept as adequate to support a conclusion.11
B. Substantial Evidence Supports the Board’s Decision.
The Alaska Workers’ Compensation Act creates a presumption that an employee’s claims are compensable.12 Applying this presumption involves a three-step analysis.13 First, to trigger the compensability presumption the employee must establish a link between his injury and his employment.14 In this case, the board found that Orchitt had produced sufficient evidence to establish a link between Orchitt’s employment and his disability. AT & T does not appear to contest this part of the board’s findings.
Second, once the employee establishes the presumption of compensability, the employer may rebut the presumption with substantial evidence.15 In Orchitt’s case, the board found that AT & T had rebutted the presumption. Orchitt does not challenge the board’s finding that AT & T rebutted the presumption.
Third, if an employer rebuts the presumption, the burden shifts to the employee to prove his claim by a preponderance of the evidence.16 Here the board found that Or-chitt had provided sufficient evidence to establish his claim. AT & T challenges the board’s conclusion that Orchitt proved his claim by a preponderance of the evidence, raising several issues about whether substantial evidence supports the board’s decision.
1. Overexposure to radio frequency radiation
AT & T first claims that the board’s finding that Orchitt was overexposed to radio frequency radiation is not supported by substantial evidence because the board rejected the testimony of the board’s engineering expert and AT & T’s radiation expert and relied instead on the lay testimony of Orchitt and his coworker, as well as its own common sense. AT & T asserts that Dr. Ziskin, a medical doctor who also calculated Orchitt’s exposure, retracted his initial opinion that Orchitt had been overexposed. It argues that because the issue of overexposure to radio frequency radiation is highly technical, any finding that Orchitt was overexposed must be supported by expert scientific testimony. It alleges that only Kimberly Kant-ner and Dr. Guy had adequate expertise to properly evaluate the level of Orchitt’s exposure.
In some workers’ compensation cases expert medical testimony is necessary to demonstrate a relationship between the claimant’s employment and his disability.17 Whether expert testimony is necessary depends on the probative value of the available lay evidence and the complexity of the medical facts involved.18 AT & T relies on Commercial Union Cos. v. Smallwood in arguing that the board erred in its finding of overexposure. But the board based its finding that Orchitt was overexposed to radiation not just on lay testimony and common sense; it also relied on Dr. Ziskin’s expert opinion. Although AT & T contends that Dr. Ziskin retracted his opinion, the record does not support this assertion. In his April 16, 1999 letter, he did not, as AT & T argues, say that it was unlikely that Orchitt sustained any significant overexposure. This letter contains no reference at all to exposure level. Nor did Dr. Ziskin change his mind in his deposition. Dr. Ziskin there testified that he still had concerns about whether Dr. Guy’s models adequately accounted for “specular reflection” in determining how quickly the radiation would dissipate once it left the wav*1241eguide. AT & T points to the following excerpts from Dr. Ziskin’s deposition testimony to support its argument:
Q: Okay. All right. You haven’t done any calculations or analysis yourself, have you, to the degree Dr. Guy has?
A: About what? I mean, I do make calculations of radiation exposures and so on from antennas and things like that.
Q: No. I mean in this case, Doctor. I’m sorry.
A: Oh, in this particular case?
Q: Yeah.
A: The calculations that I made were very—
Q: Rough?
A: —limited. I took the total power that was coming through the waveguide, and I divided it by the area to come up with what was the average power density within the waveguide, which would be at the starting point.
Q: Right. And it would be — it would lose power as it moved away, right?
A: Well, within the waveguide, for the most part, the power will stay the same.
Q: Within the waveguide, but once it—
A: Within the waveguide. But once it leaves the waveguide, it gets attenuated, yeah. It depends upon the nature of the way it leaves how rapidly it attenuates.
Q: Okay.
A: And most of the modeling was done on the idea of the inverse square law; but with specular reflection, that would not necessarily be true.
Q: And specular reflection you deal with in ultrasound, right?
A: That’s correct.
Q:.... Now, as to the differences between the sort[ ] of doctor! ] that you are as compared to Dr. Guy, can you explain the differences for — so that we can understand the kind of testimonies that we can expect that you would be able to testify to accurately as opposed to the type of testimony Dr. Guy would be able to testify to accurately?
A: Well, there is a great deal of overlap. However, I’ll — I think he would defer to me when it comes to medical judgment and biology. And unless there was something very specific, I will always defer to him when it comes to the physical engineering side of things. And I think the same thing is true with — it’s possible that, because he has done some biological research, that there could be something that I would not be correct on and he would maybe correct me when it comes to even biology or even medicine, but in general, he would defer to my opinion when it comes to medical aspects.
We do not believe that Dr. Ziskin’s testimony that he would defer to Dr. Guy’s opinions with respect to physical engineering “unless there was something very specific” indicates that he retracted or otherwise abandoned his opinion that Orchitt was overexposed. Dr. Ziskin identified the specific issue of “specular reflection” as an area of possible disagreement with Dr. Guy. In his deposition, Dr. Ziskin reiterated his belief that Dr. Guy’s model had not taken “specular reflection” into account.
A: Where I have a question is in the initial assumptions of what was the exposure ahead of time, what went into that model. And that’s why I said I wanted to look to see what was the incident power density that he felt was started to expose the head with. And here is where I have a little different point of view. It has to do -with reflections off of the flange. Because I have a background in ultrasound for diagnosis where we look at reflections, that’s the whole diagnosis concerned with, I’m aware that you can have pretty large reflections off of structures that are relatively strong compared to just the scatter and the back scatter that you would have otherwise. And see, I haven’t seen the actual setup.
But it would be possible that if the two flanges that had been — that the wave*1242guide segments that had been separated had overlapped and there was strong reflections coming off of one of the flanges, that that reflection could actually be quite high and might not be measured in the model — the modeling that Dr. Guy had used. That’s sort of a rather important point because that would establish what that initial maximum exposure would be.
Q: And Dr. Guy did take that into account in the third report. Do you see that?
A: Well, looking at it, though, it’s not clear to me whether or not it actually addressed what I call specular reflections, the type of reflections I’m talking about. It looked more of the de-fraction type of reflection, which is certainly true, but I don’t know whether or not the model actually takes into account the specular reflections.
A: The only thing, I’m not sure whether that model takes into account specular reflections or not. I just don’t know for sure. I would have to ask Dr. Guy.
Dr. Ziskin’s report and testimony provide substantial scientific evidence to support the board’s finding. AT & T does not argue that Dr. Ziskin was not qualified to give an opinion about overexposure. Moreover, the board was free to credit the testimony of Dr. Ziskin over that of Dr. Guy and Kimberly Kantner. “[I]f the Board is faced with two or more conflicting medical opinions — each of which constitutes substantial evidence — and elects to rely upon one opinion rather than the other, we will affirm the Board’s decision.”19 This is particularly so if the board believed that, based on Orchitt’s description of the separation between the segments of the waveguide and his distance from the flange, specular reflection had occurred, and if it found that Dr. Ziskin’s opinion more accurately matched how the accident happened than Dr. Guy’s. We therefore find no merit in AT & T’s contention that the board’s finding was not supported by adequate scientific evidence.
Furthermore, in Beauchamp v. Employers Liability Assurance Corp., we held that the board could permissibly combine uncontra-dicted lay testimony with uncertain medical testimony to support a conclusion that a worker’s injury was work related.20 Here, the board did not err in relying on the lay testimony of Orchitt’s coworkers in combination with the medical evidence in determining that Orchitt had suffered a work-related injury.
2. Medical evidence
AT & T also argues that the board’s findings that Orchitt suffered a work-related injury and that Orchitt’s mental deficits were related to the radio frequency radiation are based on “incompetent” medical evidence. It contends that the evidence presented by Drs. Russell and Amen does not meet the standards articulated in State v. Coon21 to test the reliability of scientific testimony. Thus, it argues that the SPECT scan that was the foundation of Dr. Amen’s diagnosis of brain damage does not satisfy Coon and that because Dr. Russell was not a licensed clinical psychologist, her opinions were not sufficiently reliable to provide a basis for the board’s ruling.
In State v. Coon we set out factors for trial courts to use in determining whether expert scientific evidence is sufficiently reliable to be admitted into evidence.22 AT & T did not object before the board to the admission of either Dr. Russell’s or Dr. Amen’s reports or testimony on this basis,23 nor did *1243it make an argument about the applicability of the Coon standard to workers’ compensation cases in its superior court appeal. Because AT & T first raises the issue before us, it has waived the issue.24
AT & T did argue before the board, as it argues here, that Dr. Russell’s testimony should not be credited because she was not licensed as a clinical psychologist. But AT & T does not dispute that Dr. Russell has a doctorate degree in psychology and is certified as a rehabilitation specialist. These credentials provide her with some expertise. The board did not make a specific finding about Dr. Russell’s credibility or the weight it accorded her testimony;" nonetheless, the board acted within its discretion in rejecting AT & T’s challenge to her expertise and in admitting her testimony.25
AT & T also argues that the board must have relied on the opinions of Drs. Russell, May, and Keene in finding that Orchitt’s injury caused his impairments.26 It argues that none of these experts had sufficient expertise in radio frequency radiation exposure to be able to connect Orehitt’s injury and his medical condition.
AT & T’s argument overlooks the opinions of Drs. Ziskin and Smith. Dr. Ziskin stated in his letter to the claims adjuster that neurological problems would be expected to result from Orchitt’s overexposure to radio frequency radiation. As we have already noted, AT & T is incorrect in asserting that Dr. Ziskin withdrew his opinion about Or-chitt’s overexposure. In addition, Dr. Smith wrote that he believed that Orchitt sustained neurocognitive deficits related to radio frequency radiation. AT & T does not argue that the medical opinions of Drs. Smith or Ziskin are suspect. The board has the sole power to determine witness credibility and assign weight to medical testimony.27 “When medical experts disagree about the cause of an employee’s injury, we have held that as a general rule ‘it is undeniably the province of the Board and not this court to decide who to believe and who to distrust.’”28 Substantial medical evidence in the record supports the board’s determination that Orchitt is entitled to medical and TTD benefits.
C. The. Board Did Not Deny AT & T Due Process.
AT & T argues that a series of board procedural decisions violated its due process rights. It contends that the violations occurred when the board: (1) denied AT & T’s right to cross-examine the employee’s experts; (2) admitted Orchitt’s last-minute expert reports without giving AT & T an opportunity to rebut the evidence through an EME; (3) denied AT & T’s request for a continuance; and (4) violated its right to an impartial trier of fact when the hearing officer failed to disclose that he was also an officer of the AFL-CIO. Because we conclude that the board committed no procedural errors in these regards, it did not deny AT & T due process.
1. Cross-examination
AT & T argues that the board denied its right to cross-examine two of Orchitt’s experts, Drs. May and Keene. Dr. Keene was not available to testify at the hearing because he was out of state. Orchitt did not present Dr. May for cross-examination because Or-chitt ran out of time in presenting his case. The board gave AT & T the option of cross-examining Dr. May during the time allotted for its case-in-chief. AT & T objected to this option; in response, the board chair said, “If you want to cross-examine him, you can cross-examine him on your time tomorrow.” AT & T’s attorney indicated that she wanted to think about it; she also indicated that she wanted the record to close the next day and did not want to leave the record open for *1244depositions. The next day, AT & T’s attorney did not ask to cross-examine Dr. May, and at the end of the hearing, AT & T’s counsel explicitly stated on the record that AT & T was waiving its right to cross-examine Drs. May and Keene.
AT & T contends on appeal that this was not a true waiver because the manner in which the board proposed to permit cross-examination was constitutionally defective. It argues that the board’s admission of Dr. Keene’s report after it had been informed that Dr. Keene would not be available for cross-examination at the hearing violated Commercial Union Cos. v. Smallwood.29 In that case, we recognized that a party has the right to cross-examine a witness without bearing the cost of calling that witness herself.30 Thus, when a party files a medical report with the board, that party has the responsibility of producing the report’s author at a hearing or deposition to give the opposing party an opportunity to cross-examine the author if cross-examination is requested.31 Workers’ compensation regulations require the party seeking to introduce a witness’s testimony by deposition to pay the initial cost of the deposition.32 If a subpoena is required, the party requesting the subpoena must bear that cost as well.33 The board’s rulings here appear contrary to Smallwood because the board admitted Dr. Keene’s and Dr. May’s reports and then would have required AT & T to conduct depositions of Orchitt’s experts in order to cross-examine them.
But AT & T did not object on the record to the method of cross-examination proposed by the board here, namely deposition testimony. Instead, it simply stated that it waived its right to cross-examine Drs. May and Keene. Because AT & T did not qualify or limit its waiver of its right of cross-examination, it cannot now claim that the board erred in denying its right to cross-examine Drs. May and Keene.34
2. The board’s refusal to require an EME before the hearing
We review an agency’s application of a statute or regulation to a particular factual situation for abuse of discretion or arbitrariness.35 The board did not abuse its discretion by denying AT & T’s March 2003 pre-hearing request for a follow-up EME. Alaska Statute 23.30.095(e) provides that a medical examination requested by the employer “not less than 14 days after injury, and every 60 days thereafter, shall be presumed to be reasonable.” AT & T made its request for a follow-up EME within the time limits set out in that statute. Before making its March 2003 request, AT & T made its last request in September 2000 that Orchitt attend a medical examination. Due to Or-chitt’s new expert reports, AT & T scheduled a follow-up EME in early April 2003. The board decided that the statutory presumption for an EME was overcome because AT & T requested the follow-up EME too close to the April 8, 2003 hearing date. At the April 1 hearing, the board gave AT & T the option of obtaining a follow-up EME after the hearing. The board later ruled that after the hearing ended it would not leave the record open for AT & T to submit a follow-up EME.
Although it may appear that the board reversed course, AT & T told the board on April 8 that it wanted the record to close the following day, April 9. Because AT & T affirmatively asked that the record close on April 9, there was no reason for the board to leave the record open for AT & T to submit a follow-up EME. We cannot determine whether AT & T could have been harmed by the board’s action in denying AT & T’s request for a pre-hearing follow-up EME, because *1245AT & T apparently took no action after the hearing to obtain a follow-up EME. The board had given AT & T an opportunity to obtain a post-hearing EME. Nothing prevented AT & T from scheduling an EME after the hearing and petitioning the board to reopen the record to consider it.36 If the board had then refused to reopen the record to consider the EME, the EME would have functioned like an offer of proof available to any appellate tribunal determining whether AT & T was harmed by the board’s failure to require an EME before the hearing or its refusal to consider any evidence produced by the EME.37 And if the board had reopened the record and considered the EME evidence, any possible error in failing to require a pre-hearing EME would have been harmless.
Moreover, even though AT & T makes much of the board’s denial of a pre-hearing follow-up EME, it does not explain why other measures short of an EME would have been unsuccessful in rebutting Orchitt’s last-minute experts. AT & T does not explain, for example, why it could not have called or why it did not call Dr. Swanson, the ophthalmologist who examined Orchitt and found nothing wrong, as a witness to rebut Dr. Keene’s report. It also does not explain why cross-examination without an EME might not have been effective. AT & T also does not explain why it needed an actual examination of Or-chitt when it could have used the raw data generated by Dr. May’s tests of Orchitt.38 Furthermore, Dr. Robinson, one of AT & T’s experts, testified at the hearing that he had read Dr. May’s report, and he offered general testimony tending to discount neuropsy-chological testing.
Finally, AT & T did not object at the end of the hearing to closing the record. It did not ask to present rebuttal evidence in any form other than a follow-up EME, nor did it make an offer of proof about what evidence it might have offered in rebuttal. A party’s failure to make an offer of proof acts as a waiver of any claim of error regarding the exclusion of unspecified evidence.39
3. . The board’s denial of AT & T’s request for a continuance
Soon after Orchitt presented his new expert reports in the month before the hearing, AT & T requested a continuance of its expert medical testimony. AT & T’s continuance requests were all related to obtaining a follow-up EME to develop rebuttal evidence in response to Orchitt’s experts, Dr. May and, later, Dr. Keene. AT & T made its requests in reliance on 8 AAC 45.074.
The regulatory standard for granting a continuance is good cause.40 AT & T argued to the board that there was good cause for a continuance under 8 AAC 45.074(b)(l)(I) and 8 AAC 45.074(b)(l)(L), which state that good cause for a continuance exists when
(I) the board determines that despite a party’s due diligence in completing discovery before requesting a hearing and despite a party’s good faith belief that the party was fully prepared for the hearing, evidence was obtained by the opposing party after the request for hearing was filed which is or will be offered at the hearing, and due process required the party requesting the hearing be given an opportunity to obtain rebuttal evidence;
(L) the board determines that despite a party’s due diligence, irreparable harm may result from a failure to grant the requested continuance or cancel the hearing.
*1246“The scope of review for an agency’s application of its own regulations to the facts is limited to whether the agency’s decision was arbitrary, unreasonable, or an abuse of discretion.”41 The board did not abuse its discretion in failing to grant AT & T’s request for a continuance. The board appears to have balanced its desire to go forward with the hearing in the case, which had been pending for quite some time,42 with AT & T’s due process rights when it: (1) offered to leave the record open so that AT & T could rebut Dr. Keene’s report or cross-examine Dr. Keene at deposition;43 (2) afforded AT & T the opportunity to cross-examine Dr. May at the hearing — albeit on AT & T’s own time; and (8) offered to leave the record open at the close of the hearing.44 AT & T waived cross-examination of Drs. Keene and May and did not object to the board’s closing the record at the end of the hearing. Because the board offered AT & T some opportunity to present evidence after the hearing in lieu of granting a continuance, we cannot say that the board abused its discretion here.
4. AT & T’s right to an impartial tribunal
Due process gives a party the right to have an impartial tribunal hear the party’s case.45 AT & T contends that the hearing officer in this case was biased because he had been elected to an officer position in the Alaska Chapter of the AFL-CIO the summer before the hearing. It argues that AS 23.30.005(a) and (e) require that a workers’ compensation hearing panel be balanced, and that the panel here did not meet this requirement. It also asserts that the hearing officer should have disqualified himself under AS 44.62.450(c), one of the provisions of the Alaska Administrative Procedure Act.46 Finally, it contends that the hearing officer’s conduct violated the Alaska Code of Judicial Conduct.
a. Actual bias or prejudgment
Administrative agency personnel are presumed to be honest and impartial until a party shows actual bias or prejudgment.47 To show hearing officer bias, a party must show that the hearing officer had a predisposition to find against a party or that the hearing officer interfered with the orderly presentation of the evidence.48 We conclude that the hearing officer’s position as an AFL-CIO vice president is insufficient to show actual or probable bias on its own. Although the chair ruled against AT & T on some procedural questions, that alone is not sufficient to show a predisposition to find against AT & T. AT & T has made no showing that the hearing officer prejudged any facts in this case or was motivated by actual bias in ruling on procedural issues.
b. Workers’ compensation statute
AT & T alleges that the hearing panel violated the statutory requirement of a balanced hearing panel because the chair’s union activities upset the balance in the panel’s composition. The workers’ compensation act provides for panels of three members: a representative of labor, a representative of industry, and the commissioner of labor or *1247“the designated representative of the commissioner.”49 The applicable statute does not say that the panel must be neutral, nor does it restrict in any way whom the commissioner can appoint as his representative. There is no indication that the chair (the commissioner’s designee) was acting as a second representative of labor or in a non-neutral capacity. We are unconvinced that his ancillary union position unbalanced the panel.
c. The Alaska Code of Judicial Conduct
AT & T argues at length that the hearing officer violated the Alaska Code of Judicial Conduct but does not address the threshold issue of the code’s applicability to workers’ compensation hearing officers. It relies on one 1988 board ethics bulletin that looked to the Code of Judicial Conduct for guidance on the issue of giving references. Louisiana Pacific Corp. v. Koons, cited by AT & T to support its argument, deals with a hearing officer’s ex parte communications, which are explicitly prohibited by the Alaska Administrative Procedure Act, and says nothing about the Code of Judicial Conduct.50 Because AT & T has not adequately briefed the issue of the applicability of the Code of Judicial Conduct to workers’ compensation hearing officers, we will not consider it.51 Nor will we consider any claim that the hearing officer’s conduct violated the Administrative Procedure Act’s provision regarding disqualification of hearing officers.52
We do not believe that the hearing officer’s position as a union officer violated the code in any event. While the Code of Judicial Conduct prohibits judges from serving as officers of organizations that are likely to be engaged in proceedings that come before the judge,53 unions are not generally parties before the workers’ compensation board, even though their individual members may come before the board. Hearing officers in the workers’ compensation division are members of the Alaska State Employee’s Association, which is affiliated with the AFL-CIO. Because the hearing officer’s position as a union officer seems to have arisen directly out of his employment for the state, AT & T’s argument could potentially disqualify all hearing officers.
IY. CONCLUSION
The board did not abuse its discretion in its procedural rulings; it therefore did not deny AT & T due process. Because substantial evidence exists in the record to support the board’s findings, we AFFIRM the superi- or court judgment that affirmed the board’s rulings.