¶1 The city of Bellevue (City) appeals the superior court’s judgment reversing the Board of Industrial Insurance Appeals (Board). The Board reversed a Department of Labor and Industries (Department) award of benefits to firefighter Wilfred Larson under RCW 51.32.185 for his malignant melanoma, a form of skin cancer. The City challenges the trial court’s jury instructions, its rulings about expert testimony, and the sufficiency of the evidence to support the jury’s verdict. It also claims that the trial court should not have awarded Larson attorney fees.

¶2 Because the trial court’s instructions correctly allocated the burdens of production and persuasion as required by RCW 51.32.185(l)’s rebuttable presumption and otherwise correctly stated the law, did not mislead the jury, and allowed each party to argue its case, we reject the City’s instructional error claims. Because the City fails to demonstrate any abuse of discretion in the trial court’s challenged evidentiary rulings, we affirm those rulings. And because RCW 51.32.185(7)(b) provides for a prevailing claimant’s recovery of “all reasonable costs of the appeal” to “any court,” the court properly awarded Larson attorney fees incurred at both the trial and Board levels.

FACTS

¶3 In 2009, Wilfred Larson was diagnosed with malignant melanoma on his low back. Larson worked as a firefighter and emergency medical technician for the City since 1979. The Department allowed his claim for benefits, finding his melanoma to be an occupational disease under RCW 51.32.185(1). The City appealed to the Board, and an industrial appeals judge dismissed Larson’s claim on the City’s summary judgment motion. Larson sought review of this decision by the Board, which reversed the judge and remanded for a hearing. After a hearing, the judge issued a proposed decision and order finding that “Wilfred Larson’s condition, diagnosed as melanoma, did not arise naturally *863and proximately out of the distinctive conditions of his employment with the City of Bellevue Fire Department.” The Board denied Larson’s petition for review, and the judge’s decision became a final decision and order.

¶4 Larson appealed to superior court. The testimony in the Board record not stricken by the trial court was read to a jury. Larson presented the testimony of his witnesses first.

¶5 Larson testified about exposure at work to smoke, fumes, and toxic substances until 2010, when he was transferred to the training division. On cross-examination, he acknowledged that he engaged in outdoor activities in the summer and sometimes would not wear a shirt. He testified that he occasionally tanned in a tanning bed to prepare for summer trips in order to avoid sunburn. The City elicited testimony from Larson’s wife that the Larson family took yearly trips to Lake Chelan in the summers, that her husband would get a little pink in the sun, and that he had freckles, green eyes, and light brown hair.

¶6 Over objection, the jury heard testimony from Larson’s expert witness, Dr. Kenneth Coleman, who is board certified in family practice. He testified that Larson’s firefighter work probably was one cause of his malignant melanoma.

¶7 The jury then was read board proceeding testimony from the City’s witnesses. Dr. Andy Chien, a dermatologist and melanoma expert, testified that ultraviolet (UV) light exposure and genetic factors cause melanoma and that those with fair skin and red- or blonde-colored hair and light eyes have the highest risk for developing melanoma. He testified that UV light from sun and tanning beds is a known carcinogen for the skin. Dr. Chien testified that for male patients melanoma most often occurs on the low back. He testified that sun exposure on cloudy days, intermittent high-intensity sun exposure, or sun exposure without a blistering sunburn can lead to melanoma. He also testified that Larson probably developed melanoma from recre*864ational UV exposure and genetic risk factors. He testified that he has not seen data showing an increased occupational risk of melanoma for firefighters.

¶8 Dr. Noel Weiss, an epidemiologist, testified about his familiarity with relevant medical literature addressing any connection between firefighters’ occupational exposure and melanoma. He testified that while these studies showed higher incidences of skin cancer in firefighters, the results are unreliable to show that firefighters experience higher incidence of melanoma than the general population.

¶9 Dr. Sarah Dick, a dermatologist, treated Larson’s melanoma. She testified that it was her “highest suspicion” that UV exposure caused Larson’s melanoma. She testified that as a fair-skinned redhead, Larson belonged to a population having a higher risk of melanoma. She opined that if he had not worked as a firefighter, he could have developed melanoma. She also testified that she never advised him to stop working as a firefighter.

¶10 The Board and the trial court both excluded, as cumulative, testimony from Dr. John Hackett, a dermatologist who had examined Larson for the City.

¶11 The City, at the end of the testimony, asked the trial court to rule, as a matter of law, that the City had established by a preponderance of the evidence that Larson’s work as a firefighter was not a cause of his melanoma. According to the City, this would leave one issue for the jury to decide: did Larson prove that his melanoma was an occupational disease? The trial court denied the City’s motion.

¶12 Over the objection of the City, the trial court submitted to the jury this special verdict form:

QUESTION 1: Was the Board of Industrial Insurance Appeals correct in deciding that the employer rebutted, by a preponderance of the evidence, the presumption that Plaintiff’s malignant melanoma was an occupational disease?

ANSWER:_(Write “yes” or “no”)

*865(INSTRUCTION: If you answered “no” to Question 1, do not answer any further questions. If you answered “yes” to Question 1, answer Question 2.)

QUESTION 2: Was the Board of Industrial Insurance Appeals correct in deciding that the Plaintiff did not prove by a preponderance of the evidence that his malignant melanoma was an occupational disease?

ANSWER:_(Write “yes” or “no”)

¶13 The jury answered “No” to the first question and did not proceed further. The trial court entered a judgment in Larson’s favor that also awarded Larson attorney fees and costs incurred before the Board and the court. The City appeals. The Department has filed a respondent’s brief supporting the City’s position on the operation of RCW 51.32.185(1).

STANDARD OF REVIEW

¶14 In appeals under RCW 51.52.110, this court reviews the trial court’s factual findings for substantial evidence and if the court’s conclusions of law flow from those findings.1 At the superior court, the Board’s decision is prima facie correct. To prevail in the superior court, the appellant must establish otherwise by a preponderance of the evidence.2 This court reviews the adequacy of jury instructions de novo.3 Courts liberally construe the Industrial Insurance Act, Title 51 RCW, resolving doubts in favor of the worker.4

*866ANALYSIS

¶15 This case presents six issues for this court to resolve:

1. What are the parties’ burdens of production and persuasion when the rebuttable presumption of RCW 51.32.185(1) applies?

2. Did the trial court correctly instruct the jury about RCW 51.32.185(1)?

3. Does substantial evidence support the jury’s finding that the Board incorrectly decided that the City had rebutted by a preponderance of the evidence the presumption that Larson’s melanoma was an occupational disease?

4. Did the trial court abuse its discretion when admitting and excluding expert opinion testimony?

5. Did the trial court err by refusing to instruct the jury that it should give the testimony of Larson’s treating physician, Sarah Dick, MD, special consideration?

6. Does RCW 51.32.185(7)(b) authorize an award of attorney fees and costs incurred before the Board to a claimant who loses before the Board but prevails in superior court?

We address these issues in the order listed.

RCW 51.32.185(1) Presumption

¶16 We must decide how the presumption created by RCW 51.32.185(1) affects the burden of proof in this case. The burden of proof has two parts: a burden of production and a burden of persuasion.5 A party with the burden of production on a particular fact in issue must produce sufficient evidence to warrant submitting that issue to the trier of fact.6 Whether a party has produced a *867sufficient quantity of evidence to submit the factual issue to the trier of fact presents a legal question to be decided by the court.7

¶17 A party with the burden of persuasion must convince the trier of fact that the contested fact is true to a certain degree, using one of these three levels of persuasion: (1) beyond a reasonable doubt, (2) clear, cogent, and convincing, or (3) preponderance of the evidence.8 It comes into play when each party has met its burden of production and all of the evidence has been introduced.9 It requires that the trier of fact weigh the credibility of witnesses and the persuasiveness of the evidence.

¶18 RCW 51.32.185(1) contains a presumption that applies when a firefighter alleges certain occupational diseases:

In the case of firefighters . . . , there shall exist a prima facie presumption that: (a) Respiratory disease; (b) any heart problems, experienced within seventy-two hours of exposure to smoke, fumes, or toxic substances, or experienced within twenty-four hours of strenuous physical exertion due to firefighting activities; (c) cancer; and (d) infectious diseases are occupational diseases under RCW 51.08.140.

¶19 This statute also provides a way to rebut this presumption: “This presumption of occupational disease may be rebutted by a preponderance of the evidence. Such evidence may include, but is not limited to, use of tobacco products, physical fitness and weight, lifestyle, hereditary factors, and exposure from other employment or nonemployment activities.”10

¶20 The parties disagree about how this presumption works. Although they do not identify them by name, each *868advocates a version of the two principal competing theories of presumptions, the Thayer theory and the Morgan theory. “Under the Thayer theory, a presumption places the burden of production of evidence on the party against whom it operates but disappears if that party produces contrary evidence.”11 Under the Morgan theory, a presumption shifts the burden of proof as to the presumed fact.12 Under the Morgan theory, a presumption does not disappear upon the production of contrary evidence but continues throughout the trial, and the court instructs the jury that the party against whom the presumption operates has the burden of proving that the presumed fact is not true or does not exist.13

¶21 The City claims that RCW 51.32.185(1) creates a Thayer-like presumption. The City contends that if a firefighter shows that he has a qualifying disease, in this case melanoma, the presumption relieves him from presenting competent medical evidence relating that melanoma to his work duties to establish that it is an occupational disease. Instead, the condition is presumed to be an occupational disease, and at this point in the production of evidence, the firefighter has proved a prima facie case for relief.

¶22 According to the City, RCW 51.32.185(1) then shifts the burden of production to the employer to show by a preponderance of the evidence that the melanoma is not an occupational disease. The City contends that whether the employer has met this burden presents a question of law. If the Board decides that the employer has met this burden of production, the presumption disappears. The firefighter continues to have a burden of persuasion that requires him to establish by a preponderance of the evidence that his *869condition is an occupational disease, now without the benefit of the statutory presumption.

¶23 Under the City’s interpretation of RCW 51.32-.185(1), because the Board found that Larson’s melanoma “did not arise naturally and proximately out of the distinctive conditions of his employment,” the City claims that the Board must first have decided as a matter of law that the City rebutted the occupational disease presumption by a preponderance of the evidence. Otherwise, the Board would never have made the quoted factual finding. Therefore, the City contends that when Larson appealed to the superior court, the court should have reviewed the Board’s legal conclusion that the City had rebutted the presumption as a question of law to be decided by the court and should not have submitted it to the jury to decide as a question of fact. Instead, the trial court should have submitted to the jury as a question of fact whether Larson had established by a preponderance of the evidence that his melanoma was an occupational disease.

¶24 Larson asserts that the Morgan theory applies. Larson claims that the RCW 51.32.185(1) presumption shifts to the employer both the burden of production and the burden of persuasion. Once a firefighter shows that he suffers from a qualifying disease, RCW 51.32.185(1) establishes a presumption that the condition is an occupational disease and imposes on the employer the burden of producing evidence sufficient to establish otherwise by a preponderance of the evidence. Larson contends that whether the employer has produced this preponderance of evidence presents a question of fact. Therefore, Larson asserts that the Board weighed the conflicting evidence and answered a question of fact when it decided that the City had rebutted the presumption. When Larson appealed to the superior court, he was entitled to have a jury decide this question of fact.

¶25 Unlike many states, Washington has not adopted an analogue to Fed. R. Evid. 301, addressing the use of pre*870sumptions.14 “Washington cases apply the Thayer theory to some, but not all, presumptions and provide no general rule about when it applies.”15 Other Washington cases identify presumptions that shift the burden of proof.16 The Washington Pattern Jury Instructions (WPI) recognize Washington cases treat different presumptions differently.17 Washington case law, the text of RCW 51.32.185(1), the public policy motivating the legislature to adopt this statute, and decisions of other jurisdictions convince us that the trial court properly allowed the jury to decide if the City had rebutted the statutory presumption.

¶26 The Washington Supreme Court has approved allowing a jury to decide if the record includes sufficient evidence to overcome a presumption.18 In Luna de la Peunte v. Seattle Times Co.,19 the court considered an appeal from a trial court’s submission to a jury whether the defendant in a libel action had rebutted a presumption that the reputation and character of the plaintiff was good. It rejected this challenge, reasoning as follows:

The difficulty, however, lies not so much with the statement of law as with its application. When and by whom is it to be said that the evidence is sufficient to rebut a presumption? In some cases, doubtless, it may be said by the court, as a matter of law. In other cases, it may very properly be left to the jury to say. Simply because some witness takes the stand and swears to facts which, if true, would rebut the presumption, does not require the court to hold, as a matter of law, that the presumption has been rebutted. . . .

*871The sum and substance of all that has been written on the force and effect of presumptions is that, in the first instance, it is for the court to say whether or not the evidence is sufficient, as a matter of law, to overcome a presumption. If not, the question may be left to the jury, under proper instruction.[20]

Thus, unless the trial court could rule as a matter of law that the City had proved by a preponderance of the evidence that Larson’s melanoma was not an occupational disease, it properly submitted this question to the jury To make this ruling, CR 50(a)(1)21 requires that “there is no legally sufficient evidentiary basis for a reasonable jury to find or to have found” for Larson on the issue. Because, as we discuss below, sufficient evidence supports the jury’s verdict in favor of Larson, the record does not meet the CR 50(a)(1) standard. As a result, we do not address whether a court could ever decide as a matter of law that the evidence produced rebutted the presumption.

¶27 The text of RCW 51.32.185(1) supports the conclusion that this statute shifts both the burden of persuasion and production. RCW 51.32.185(1) states that the “presumption of occupational disease may be rebutted by a preponderance of the evidence.” Although the statute does not define “preponderance of the evidence,” WPI 21.01 does:

When it is said that a party has the burden of proof on any proposition, or that any proposition must be proved by a *872preponderance of the evidence, or the expression “if you find” is used, it means that you must be persuaded, considering all the evidence in the case [bearing on the question], that the proposition on which that party has the burden of proof is more probably true than not true.

(Alteration in original.)

¶28 Neither party suggests a different definition, nor are we aware of any reason to use a different one. Thus, the statute requires a quality of proof to rebut the presumption and a weighing of all the evidence to determine if the evidence produced achieves the necessary level of persuasiveness. This presents a question of fact requiring an evaluation of the credibility of witnesses and the persuasiveness of evidence. Logically, this presumption shifts to the City the burden of proof as to the presumed fact of occupational disease.22

¶29 Division Two of this court appears to have reached this conclusion in Gorre v. City of Tacoma:23 “[T]he burden shifts to the employer to rebut the presumption by a preponderance of the evidence by showing that the origin or aggravator of the fire fighter’s disease did not arise naturally and proximately out of his employment.”24 The City must present evidence to “show that his diseases did not arise from his fire fighter employment.”25

¶30 A leading commentator on Washington evidence law suggests that a presumption written in terms of the level of proof required to overcome it follows the Morgan theory and *873that the WPI adopts this view.26 A note for use of one pattern instruction supports his view. WPI 24.05 provides, “[If you find] [Because] (insert the basic facts), the law presumes (insert the presumed fact), and you are bound by that presumption unless you find [by a preponderance of the evidence] [by clear, cogent, and convincing evidence] that (insert the contrary of presumption).” (Alterations in original.) This pattern instruction parallels the presumption and rebuttal provision of RCW 51.32.185(1). The note on use for this instruction states,

This instruction is proper only for rebuttable mandatory presumptions that affect the burden of proof, when the presumed fact has been challenged and constitutes a jury question. It should therefore not be given if the court can rule on the presumed fact or facts as a matter of law. The bracketed phrase “if you find” should be used if the basic facts are also a jury question. The bracketed word “because” should be used if the basic facts are admitted or can otherwise be determined to exist as a matter of law. In applying this instruction, the user should pick out the appropriate burden of proof necessary to rebut the presumption and then give a definition of that burden.

This note on use suggests that the Washington Supreme Court Committee on Jury Instructions would consider the presumption of RCW 51.32.185(1) as a rebuttable presumption that affected the burden of proof.

¶31 The Morgan theory of presumptions provides that when a presumption reflects a strong social policy, the presumption should shift the burden of persuasion.27 For this reason, a number of jurisdictions have applied the Morgan theory to statutes substantively similar to RCW 51.32.185(1).28 For example, the New Hampshire Supreme Court decided the apparent purpose of an analogous New *874Hampshire statute was to implement a social policy of providing compensation to firefighters suffering identified diseases where the medical evidence fails to establish the cause of the disease.29 To give full effect to the legislative intent, the court applied the statutory presumption “with a force consistent with the legislative concerns underlying the presumption.”30 The court decided that the Thayer theory would be inconsistent with the policy objective of the presumption and applied the Morgan theory.31 The North Dakota Supreme Court has described the purpose of North Dakota’s analogous statute as relieving firefighters of the nearly impossible burden of proving firefighting actually caused their disease.32

¶32 The Washington Legislature first adopted RCW 51-.32.185 for similar reasons:

The legislature finds that the employment of fire fighters exposes them to smoke, fumes, and toxic or chemical substances. The legislature recognizes that fire fighters as a class have a higher rate of respiratory disease than the general public. The legislature therefore finds that respiratory disease should be presumed to be occupationally related for industrial insurance purposes for fire fighters.[33]

When the legislature amended RCW 51.32.185 in 2002 to add a presumption for melanoma, it made a finding that “[a] 1990 review of fire fighter epidemiology calculated a statistically significant risk for melanoma among fire fight*875ers.”34 Our governor vetoed the bill section containing this finding.35 But this legislative history makes clear the social purpose of the presumption. We agree with the New Hampshire Supreme Court’s conclusion that the Morgan theory should be applied to the presumption to give it the force intended by the legislature.

¶33 The City and Department cite employment discrimination cases to argue that once Larson established his prima facie case, the City had only a burden of production at this stage, an issue of law not reviewable by a jury. We do not find this analogy of employment discrimination procedure to the operation of a statutory presumption persuasive. It ignores Washington case law cited above and the text of RCW 51.32.185(1).

¶34 In summary, once a firefighter proves that he suffers from a qualifying disease described in RCW 51.32-.185(1), this statute’s presumption shifts the burdens of production and persuasion to the entity contesting an award of industrial insurance benefits. The trial court did not err in allowing the jury to decide if the City had rebutted this presumption. Our conclusion comports with a liberal construction of the Industrial Insurance Act in favor of the worker.36

Jury Instructions about RCW 51.32.185(1)

¶35 The City contends that jury instruction 9 misstated the proof required to rebut RCW 51.32.185(l)’s presumption. It claims that the third paragraph of this instruction told the jury that the City had to rebut “two elements” of the definition of “occupational disease” to rebut the statutory presumption when it had to rebut only one “element.” The City argues that the special verdict form exacerbated this erroneous statement of the law because it required the jury *876to reference the flawed instruction and contained the word “presumptions,” indicating to the jury that more than one existed. It argues this prejudiced the City’s ability to argue its case theory. Larson argues that the City improperly objected to the claim precluding this court’s review and, regardless, that the instruction correctly stated the law.

¶36 Instruction 9 states,

The findings and decision of the Board of Industrial Insurance Appeals are presumed correct. This presumption is rebut-table, and it is for you to determine whether it is rebutted by the evidence.

The burden of proof is on the firefighter to establish by a preponderance of the evidence that the decision is incorrect.

At the hearing before the Board of Industrial Insurance Appeals, the burden of proof is on the employer to rebut the presumption that 1) claimant’s malignant melanoma arose naturally out of his conditions of employment as a firefighter and, 2) his employment is a proximate cause of his malignant melanoma.

When it is said that a party has the burden of proof on any proposition, or that any proposition must be proved by a preponderance of the evidence, or the expression “if you find” is used, it means that you must be persuaded, considering all the evidence in the case bearing on the question, that the proposition on which that party has the burden of proof is more probably true than not true.

¶37 Sufficient instructions permit parties to argue their case theories, do not mislead the jury, and, when read as a whole, accurately inform the jury of applicable law.37

¶38 RCW 51.08.140 defines “occupational disease”: “ ‘Occupational disease’ means such disease or infection as arises naturally and proximately out of employment under the mandatory or elective adoption provisions of this title.” Instruction 9 breaks this definition into two parts — that *877malignant melanoma arose naturally out of conditions of employment and that employment was a proximate cause of the melanoma. It directly tracks the statutory definition. We agree that the City needed to disprove only one of the two parts to rebut RCW 51.32.185(l)’s presumption. Nothing in instruction 9 contradicts this. And, contrary to the City’s assertion, the special verdict form did not contain the word “presumptions” or place an improper additional burden of proof on the City. Thus, we conclude that instruction 9 and the special verdict form correctly stated the law, did not mislead the jury, and allowed the City to argue its theory of the case.

Sufficiency of the Evidence

¶39 The City argues that substantial evidence did not support the jury’s finding that the Board incorrectly decided that the City rebutted the presumption that Larson’s melanoma was an occupational disease. A court will not overturn a jury verdict if substantial evidence exists to support it.38 “Under the substantial evidence standard, there must be a sufficient quantum of evidence in the record to persuade a reasonable person that the declared premise is true.”39

¶40 We must decide if Larson presented substantial evidence that rebuts the presumption that the Board correctly decided the case.40 The Board heard testimony from three of the City’s medical expert witnesses. Through these witnesses, the City elicited expert and lay testimony showing that UV rays from the sun and tanning beds as well as genetic factors can cause melanoma. This testimony suggested that Larson had a higher risk for melanoma for reasons not related to his occupation. Dr. Chien, in particu*878lar, identified recreational UV exposure and genetic risk factors as the cause of melanoma diagnosed in a firefighter with Larson’s history. Epidemiologist Dr. Weiss, testifying for the City, opined that in his professional opinion the studies showing firefighters experience higher incidence of melanoma produced unreliable results. And Dr. Dick, Larson’s treating dermatologist, testified that it was her “highest suspicion” that UV exposure caused his melanoma. She testified that Larson could have developed melanoma even if he had not been a firefighter.

¶41 But Larson points to evidence in the record demonstrating that a reasonable juror could find flaws in the City’s case. Dr. Chien testified, for example, that it is unknown what quantity or dose of UV exposure one requires to develop melanoma and that the exact cause of melanoma can never be determined. He testified on cross-examination that “if you look at whether or not asking about someone’s chronic sun exposure can predict their risk for melanoma, the studies so far have said that it cannot.” Larson elicited testimony to show the jury that Dr. Chien does not know the time or circumstance when a person becomes positive for melanoma and that he had not examined Larson. Dr. Chien acknowledged that he was not familiar enough with what a firefighter does to know if he or she is exposed to ionizing radiation, a form of UV exposure that he testified can be a cause of melanoma.

¶42 Larson asked Dr. Weiss, the City’s witness epidemiologist, about a 2007 “Registry Based Case Control Study of Cancer in California Firefighters.” Dr. Weiss testified, “In this study, they did find a suggestion of [a] 50 percent increase in risk[ ] of melanoma among firefighters.” He identified a potential design flaw, testifying that the study’s “results are helpful” because the study added to all of the studies but that it is one small piece of evidence. When asked about a bias for UV exposure in this study, Weiss testified that he believes that bias “would not be terribly large.” He testified, “[L]et’s say all of the studies collectively *879found a 50 percent increase. I wouldn’t think that the potential difference in sun exposure would be responsible for all of that difference.”

¶43 Dr. Dick testified that to her knowledge there is no way to look at a biopsy or the melanoma and determine what caused it and that there can be more than one cause to one’s malignant melanoma. When the City asked her medical opinion about whether Larson would have developed skin cancer if he didn’t work as a firefighter, she answered, “At this point I don’t know if I can give a medical opinion because I don’t know enough details of what he does at work.” When pressed, she continued, “[N]ot knowing his exposure to UV during his work, assuming he’s covered at work, then do I think — do I think he would have gotten skin cancer if he wasn’t a firefighter? Yeah, he could have.”

¶44 While the City elicited expert testimony to highlight Larson’s nonoccupational risks for developing melanoma, a jury could reasonably conclude from the testimony of these witnesses that the City had not disproved firefighting as a more probable than not cause for Larson’s melanoma. Thus, substantial evidence existed to support a finding that the Board incorrectly concluded the City rebutted its presumption.

Expert Testimony Decisions

¶45 This court reviews a trial court’s admission or rejection of expert testimony for an abuse of discretion, looking to see if the court made a manifestly unreasonable decision or based it on untenable grounds.41

¶46 The City challenges the trial court’s admission of expert testimony from Larson’s expert, Dr. Coleman, and the court’s rejection of testimony by its own expert, Dr. Hackett. The City claims Dr. Coleman’s testimony did not qualify him as an expert under ER 702 and provided *880untrustworthy and unreliable evidence that allowed the jury to speculate about medical causation. Alternatively, it argues that the trial court should have stricken portions of his testimony because it responded to Larson’s leading questions and constituted an improper use of learned treatises.

¶47 ER 702 governs testimony by experts: “If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise.” This rule provides a two-step inquiry: does a witness qualify as an expert and will the expert testimony be helpful to the trier of fact?42 Expert testimony must be “sufficiently trustworthy and reliable to remove the danger of speculation and conjecture” from the jury.43 Courts usually admit expert testimony “under ER 702 if it will be helpful to the jury in understanding matters outside the competence of ordinary lay persons.”44

¶48 Larson’s expert witness, Dr. Coleman, testified about the content of 12 peer-reviewed scientific articles based on his “extensive review” of them for trial. He is a licensed medical doctor, is board certified in family practice medicine, and has run several hospital emergency departments. He diagnoses skin diseases, does biopsies, and recognizes changes in skin lesions, and testified to his familiarity with the causes of melanoma as part of his practice and through expert witness work on a different case. Based on his reading of the articles, he testified about the relationship between chemicals and cancer, firefighters’ occupational exposure to chemicals, increased incidence of mela*881noma in firefighters, and literature that identifies risk factors for malignant melanoma aside from sunlight. Coleman formed an opinion, based on those articles and his training and experience, that Larson’s “occupational exposure as a fire fighter must be considered here on a more-probable-than-not-basis as one of the causes for his development of malignant melanoma.”

¶49 The City cites Harris v. Robert C. Groth, MD, Inc. 45 to show that Coleman’s testimony about the cause of malignant melanoma “is outside his expertise as a family practice physician.” In that case, the court held that nonphysicians may give expert testimony but affirmed the trial court’s exclusion of expert witness testimony as within the court’s discretion.46 The court stated that “[t]rial courts retain broad discretion in determining whether an expert is qualified.”47

¶50 In this case, the trial court denied the City’s motion to strike Dr. Coleman’s testimony because “the witness clearly qualifies as an expert with information that would assist the trier of fact, and that’s based not only on the recitation of the qualifications, but also my review of the portions of the record that were contained in the briefing.” Because Dr. Coleman is a licensed physician who has experience diagnosing skin diseases, including melanoma, we conclude that he had sufficient knowledge and experience to provide useful information to the jury.

¶51 The City further challenges portions of Dr. Coleman’s testimony involving the articles as inadmissible hearsay. ER 802 bars hearsay, except as permitted by the Rules of Evidence, other court rules, or statute. ER 803(a)(18) provides that statements contained in published treatises, periodicals, or pamphlets on a subject of medicine are not hearsay “ [t] o the extent... relied upon by the expert *882witness in direct examination.” These statements may be read into evidence if the expert testifies to their reliable authority.48 And ER 703 allows experts to use facts or data reasonably relied on in the expert’s field to form an opinion or make inferences.

¶52 The City claims that on direct examination an expert may only “rely upon” treatises, and here, Larson improperly “called to [Dr. Coleman’s] attention” the articles’ contents. The City, citing no supporting authority, argues that when Dr. Coleman responded by commenting on the articles’ correctness, instead of reading the passages to the jury himself, his testimony fell outside of the hearsay exception in ER 803(a)(18).

¶53 But ER 803(a)(18) states that “[i]f admitted, the statements may be read into evidence.” The rule does not limit who may read them. Dr. Coleman identified the articles as authoritative on the subject of malignant melanoma, that medical experts reasonably rely on these peer-reviewed articles, and that he relied on these in his testimony regarding causation of malignant melanoma. Larson points to multiple examples where Dr. Coleman relied on the articles to form his own conclusions. Thus, the trial court did not abuse its discretion by allowing Dr. Coleman’s testimony.

¶54 The City next asserts that the trial court erred by excluding Dr. Hackett’s testimony as cumulative. The City argues that unlike its other expert witnesses, Dr. Hackett examined Larson with the purpose of reviewing risk factors that may have led to Larson’s melanoma and would have opined about the role of those risks in Larson’s case. For example, the City argues, the jury would have heard testimony from Dr. Hackett about Larson’s use of tanning beds, when Dr. Dick failed to ask Larson about this. But Larson testified that he used a tanning bed, and the City established through Dr. Dick’s testimony that this *883could cause his melanoma. Dr. Dick’s testimony also reviewed her examination of Larson and addressed risk factors for malignant melanoma. Because the trial court may limit the number of expert witnesses at trial and has discretion to admit cumulative evidence, expert or not,49 the court did not abuse its discretion in excluding Dr. Hackett’s testimony as cumulative.

Attending Physician Instruction

¶55 The City contends that the trial court erred when it refused to give WPI 155.13.01, relating to special consideration of an attending physician’s testimony. Courts often instruct juries to give special consideration to a claimant’s attending physician.50 Here, however, the trial court expressed concern that the instruction would be misleading. It reasoned that “it seems odd to me to direct the jury to especially consider the treating physician in a case where the testimony is, I sure wish I had my chart,... and where some of the other testimony has been, you know, more elaborate and covering.”

¶56 In Boeing Co. v. Harker-Lott,51 this court held that an instruction on special consideration of attending physicians is not mandatory and that a more general instruction still allowed a party to argue its theory of the case. Similarly, the court here instructed the jury that it may determine witness credibility by taking into consideration factors including the “opportunity of the witness to observe or know the things they testify about” and “any personal interest that the witness might have in the outcome or the issues.” This instruction allowed the City to argue that Dr. Dick’s testimony was especially helpful because she was Larson’s only attending physician who testified, was neutral, and did not advise Larson to forgo *884firefighting or take precautions at work to limit potential recurrence. Because the trial court here included instructions that properly informed the jury and allowed the City to argue its case, the court did not abuse its discretion by declining to give an instruction on special consideration of attending physician’s testimony.

Attorney Fees

¶57 The trial court awarded Larson $67,470.00 in attorney fees and $12,132.42 in costs under RCW 51.32-.185(7) and RCW 51.52.130. The City argues that the trial court should not have included in the award fees and costs Larson incurred for his unsuccessful Board appeal. Because the language of RCW 51.52.130 unambiguously provides for recovery “for services before the court only,” this statute does not authorize an award of attorney fees for unsuccessful Board appeals ultimately reversed on further appeal to superior court.52 The City argues that RCW 51.32.185(7) has the same limitation because its text divides recovery into two sections — one for successful board appeals and another for successful appeals to the courts. It argues that this precludes Larson’s recovery for his unsuccessful appeal to the Board, regardless of his success at trial.

¶58 Larson argues that RCW 51.32.185(7)(b)’s unambiguous language permits recovery. We agree with Larson. When a firefighter claimant successfully appeals at trial court, RCW 51.32.185(7)(b) provides that “the court shall order that all reasonable costs of the appeal, including attorney fees and witness fees, be paid to the firefighter or his or her beneficiary by the opposing party.” Unlike RCW 51.52.130, it does not limit recovery to “services before the court only.” We conclude that the plain language of “all reasonable costs of the appeal” includes all, and not only some, of the costs required to succeed on a claims benefit under the Industrial Insurance Act. This construction fur*885thers the purpose of awarding attorney fees in industrial insurance cases: “ ‘to guarantee the injured workman adequate legal representation in presenting his claim on appeal without the incurring of legal expense’ ” if the claimant prevailed.53 It is also consistent with our obligation to construe the Industrial Insurance Act liberally in favor of the worker. The court properly included in its fee and cost award the fees and costs Larson incurred before the Board.

CONCLUSION

¶59 The trial court’s instructions correctly allocated the burdens of production and persuasion as required by RCW 52.32.185(l)’s rebuttable presumption and otherwise correctly stated the law, did not mislead the jury, and allowed each party to argue its case. Because the City fails to demonstrate any abuse of discretion in the trial court’s challenged evidentiary rulings, we affirm those rulings. And because RCW 51.32.185(7)(b) provides for a prevailing claimant’s recovery of “all reasonable costs of the appeal” to “any court,” the court properly awarded Larson attorney fees incurred at both the trial and Board levels. We affirm the trial court.

Review granted at 184 Wn.2d 1033 (2016).