SERCOMBE, J.
After a contested case hearing, the Oregon Medical Board issued a final order revoking petitioner’s license to practice medicine and awarding costs based on its conclusion that his treatment of nine patients constituted unprofessional or dishonorable conduct, gross or repeated negligence, or a willful violation of prior board orders. Petitioner seeks judicial review of the board’s order. Among his many arguments on review, petitioner challenges the board’s legal conclusions as to his treatment of specific patients. More broadly, he contends that, contrary to the board’s conclusion, the treatment in question constitutes “alternative medical treatment,” as defined in ORS 677.190(l)(b)(A), and that it therefore cannot be deemed unprofessional. As explained below, although we largely reject petitioner’s challenges to the board’s legal conclusions, we conclude that the board erred in determining that petitioner’s treatment of two patients, I and C, violated prior board orders. Accordingly, we reverse and remand for reconsideration.
We take the following facts from the board’s final order. In 1994, the board revoked petitioner’s license to practice medicine based on its conclusion that he had engaged in unprofessional conduct. Petitioner’s license was later reinstated pursuant to a stipulated order, which included various limitations on his practice. In 2002, the board opened an investigation, and petitioner requested an amendment to the stipulated order to include additional restrictions. In 2004, in a modified stipulated order (the 2004 order), petitioner agreed to the following limitations on his use of thyroid medication:
“[Petitioner] shall use thyroid function blood tests (with appropriate documentation in the patient charts), to include blood tests that measure the amount of thyroid hormone (free T4) and thyroid-stimulating hormone (TSH) in conjunction with the history and physical findings in making the decision whether to use thyroid medication. [Petitioner] will not use thyroid medication in treatment unless the blood tests find a TSH level greater than the normal range and a free T4 below the normal range on a test. While treating patients with thyroid medication, [petitioner] shall periodically retest the TSH level of his patients’ blood [171]*171no later than six weeks after initiating treatment with thyroid medication and no less than annually thereafter. [Petitioner] shall reduce the level of thyroid medication if the level of TSH falls below the normal range.”
In September 2009, the board notified petitioner that it had opened another investigation, based on allegations from patient A that petitioner had recommended that she take thyroid, even though tests showed that her thyroid levels were normal, and that she obtain that thyroid from Mexico through an online source. As the investigation continued, petitioner proposed further restrictions on his treatment with thyroid and testosterone hormones. The board adopted those restrictions and in March 2010 issued an interim stipulated order (ISO), which modified the 2004 order by adding the following:
“3.1 [Petitioner] will not recommend, prescribe, or direct any patient to take thyroid unless patient TSH levels exceed 10 uIU/mL, except that [petitioner] may recommend, prescribe or direct a patient to take thyroid supplementation if patient TSH levels are between 5 and 10 uIU/ mL and the patient has also been diagnosed with goiter or positive anti-thyroid peroxidase antibodies (or both).
“3.2 [Petitioner] will require any patient taking thyroid from a non-prescription source to undergo thyroid blood tests on a regular basis (at least every 6 months) and that [petitioner] will direct such patients to adjust their dose to bring their TSH level into the range recommended by the American Association of Clinical Endocrinologists (AÁCE) (the target TSH level is between 0.3 and 3.0 uIU/mL). If any patient declines to follow this direction, [petitioner] will provide 30 day prior written notice to the patient and then terminate the physician-patient relationship.
“3.3 In the event [petitioner] decides to prescribe, recommend, direct a patient to take testosterone, or to follow a patient taking testosterone, [petitioner] must comply with the guidelines recommended in the article ‘Risks of Testosterone Replacement Therapy and Recommendations for Monitoring,’ published in the New England Journal of Medicine, 350:5, January 29, 2004. Specifically, [petitioner] will ensure that either he or another physician has conducted and documented a recent digital rectal examination and that at a minimum, blood tests for baseline testosterone [172]*172and PSA levels have been performed. [Petitioner] must not prescribe, recommend, or direct a patient to take testosterone for patients with a PSA level above 4.0 ng/mL as well as patients with a yearly PSA increase of 1.5 ng/mL or more, or 0.75 ng/mL per year or more over two years. If any patient insists that they want to take testosterone in the face of such PSA levels, [petitioner] will provide 30 day prior written notice to the patient and then terminate the physician-patient relationship.
“3.4 [Petitioner] must make appropriate and timely chart entries to demonstrate that he is complying with the terms of this Order.”
In May 2010, the board issued a complaint and notice of proposed disciplinary action. From a list of patients petitioner treated with either thyroid or testosterone during the prior year, a board investigator randomly selected five patient charts to review. After reviewing those charts (for patients H, I, J, K, and L), the board determined that petitioner was not in compliance with the ISO and that petitioner’s practice of medicine would pose an immediate danger to the public and to his patients. In September 2010, the board issued an emergency order suspending petitioner’s license to practice medicine.
That same month, the board issued an amended complaint and notice of proposed disciplinary action. The amended complaint alleged that petitioner had failed to comply with paragraph 3 of the ISO in treating patients H through L. The complaint further alleged that petitioner’s treatment of five other patients (patients A, B, C, D, and E) with thyroid constituted unprofessional or dishonorable conduct under ORS 677.190(l)(a); gross or repeated acts of negligence under ORS 677.190(13); and was a willful violation of the 2004 order under ORS 677.190(17).1 Specifically, [173]*173the complaint alleged that petitioner had failed to use appropriate testing to diagnose hypothyroidism, monitor patient response to treatment, and set forth a clinical basis for diagnosing and treating hypothyroidism. For two other patients, F and G, the complaint alleged that petitioner’s prescription of testosterone that was not medically indicated constituted unprofessional or dishonorable conduct and gross or repeated negligence and exposed those patients to a risk of harm. Petitioner requested a contested case hearing.
The hearing was presided over by an administrative law judge (AL J), who issued a proposed order. The AL J relied on testimony from two physicians, Cook and Nedrow, in determining the standard of care for “medical practitioners practicing within the specialization of endocrinology and hormone replacement.”2 According to Cook and Nedrow, the standard of care for treating a patient for hypothyroidism requires an initial diagnosis based upon a thyroid-stimulating hormone (TSH) blood test in addition to clinical examination and history. After an initial diagnosis, TSH levels should be checked, at a minimum, yearly, and after every dosing change, to monitor the patient’s reaction to exogenous thyroid. According to guidelines set by the American [174]*174Association of Clinical Endocrinologists (AACE) and the Endocrine Society, the “normal target range for TSH is from 0.3 uIU/mL to 3.0 uIU/mL.” A TSH above the normal range, in general, is an indication for treatment with thyroid to bring the number down to within that range. Apart from TSH testing, “testing for specific thyroid antibodies, free-T4 and free-T3, provides complimentary information for the primary diagnosis of hypothyroidism, or low production of thyroid by the thyroid gland.” Risks to a patient who receives excess thyroid include “accelerated bone density loss, or osteopenia, cardiac complications including cardiac arrhythmia, and developing hyperthyroidism, which may increase the risk for hypertension.”
The ALJ also relied on testimony from Cook and Nedrow regarding the use of testosterone to treat hypogonadism, which the board described as a “clinical condition in which low levels of serum testosterone are found in association with specific signs and symptoms, including diminished libido and sense of vitality, erectile dysfunction, reduced muscle mass and bone density, depression, and anemia.” Testosterone replacement therapy requires “[a]n accurate diagnosis of hypogonadism” based on “[a] blood test to establish a baseline testosterone serum level * * * in association with other specific signs and symptoms [.]”
The ALJ found that “ [t] estosterone replacement therapy is linked to exposing underlying cancers and to accelerating tumor growth” and that “ [a] dditional risks of having excess testosterone are prostate hypertrophy and hyperplasia, and possible urinary tract obstruction.” Although the ALJ acknowledged “controversy regarding the cause and effect relationship between testosterone replacement therapy and prostate tumor growth,” it found that “the current standard of care for allowing testosterone therapy replacement in men with a history of prostate cancer is a disease-free interval, meaning no recurrence of prostate cancer, of five years.” Treatment with testosterone for an individual with a prostate-specific androgen (PSA) level in excess of 4.0, under either the AACE or the Endocrine Society guidelines, constitutes a breach of the standard of care, and “ [treatment with testosterone therapy for men [175]*175with an abnormal DRE [digital rectal exam], elevated PSA level or rapidly increasing PSA levels is not within the standard of care, and is contra-indicated.”
Although the ALJ relied upon Cook and Nedrow to determine the standard of care in the medical community, the ALJ acknowledged the contrary opinion presented by petitioner’s medical experts: doctors Green, Mead, and Welker. Regarding diagnosis of hypothyroidism, Green testified that he relies more on a history of certain symptoms and basal temperature in diagnosing hypothyroidism, that a patient’s TSH level “is not an effective diagnostic tool for hypothyroidism,” and that he agreed with petitioner’s use of thyroid in treatment. Mead testified that, in his opinion, there is no risk associated with treating a patient with a diagnosis of prostate cancer with testosterone and that petitioner’s treatment with testosterone posed no greater risk to either patient than the generally recognized treatment. Welker generally agreed with these assessments and concluded that petitioner’s use of thyroid and testosterone was reasonable and did not increase the risk of harm over the generally recognized treatment.
In the ALJ’s view, petitioner’s experts were not as reliable as Cook or Nedrow; the ALJ found that Cook and Nedrow were highly qualified based on their education and experience and that their “opinions were supported by authoritative sources in the evidence record.” By contrast, the ALJ found that Welker was not qualified to testify as to the standard practice of endocrinologists or integrative medicine practitioners because he was trained in other areas. The ALJ also found that Welker’s and Mead’s medical judgments were less reliable than the board’s experts because Welker had been subject to disciplinary action and loss of hospital privileges and Mead had been in a business relationship with petitioner.
The ALJ issued a proposed order concluding that petitioner violated the ISO by treating Patient L for hypothyroidism without establishing thyroid levels and by treating patient I with testosterone even though he did not have an abnormal testosterone level. The ALJ concluded that petitioner’s treatment of patients H, J, and K did not violate [176]*176the ISO. In the ALJ’s view, petitioner’s treatment of patients I and L “constituted a repeated disregard of prior Board Orders, unprofessional or dishonorable conduct, and gross or repeated negligence,” and as a result, petitioner’s “conduct posed an immediate danger to the public and to his patients, and required the immediate suspension of [petitioner’s] license.”
The ALJ also found that, when treating patients A through E with thyroid, petitioner “provided treatment to patients based on the patients’ reported feelings of well-being despite there being well-known risks associated with the treatments. In addition, [petitioner] lacked diagnostic criteria to make accurate diagnoses, which is necessary in practicing the conventional medicine demonstrated in the charts of the patients at issue.” As such, the board concluded that petitioner “willfully disobeyed a Board order by failing to comply with the terms * * * of the 2004 [order] as alleged,” and his treatment “constituted unprofessional or dishonorable conduct and gross or repeated negligence that exposed his patients to harm.” As to patients F and G, the ALJ concluded that petitioner “prescribed testosterone when not medically indicated, and failed to set forth in the patient charts the clinical basis for diagnosing and treating hypogonadism. [Petitioner’s] conduct breached the standard of care, and constituted unprofessional or dishonorable conduct and gross or repeated negligence.”
In addressing petitioner’s claim that he was practicing “alternative medical treatment” under ORS 677.190(l)(b)(A), the ALJ concluded that, “[t]o the extent that the experts determined that [petitioner’s] thyroid and testosterone replacement therapies were not alternative medical treatments but were conventional treatments for conventional diagnoses treated in a manner that was outside the standard of medical care, the provisions of ORS 677.190(l)(b) do not apply.” Further, though, the ALJ specifically concluded that, even if the first two criteria of ORS 677.190(l)(b) were satisfied, “the Board met its burden to show * * * that [petitioner’s] treatment posed a greater risk of harm to his patients than the standard or recognized treatment, contrary to the provisions of ORS 677.190(l)(b)(A)(iii).”
[177]*177After considering petitioner’s exceptions to the proposed order, the board adopted the AL J’s findings of fact and conclusions of law set out above. The board revoked petitioner’s medical license and assessed the costs of the hearing.
In his first two assignments of error, petitioner presents several arguments challenging the validity of the ISO and the emergency suspension order. We reject those arguments without published discussion. In his third assignment of error, petitioner argues that the board erred in concluding that his treatment of two patients, L and I, violated the ISO and supported the sanctions imposed in the final order.3 We review the board’s conclusion for legal error. ORS 183.482(8)(a).
Before the ISO took effect, petitioner diagnosed patient L with hypothyroidism without TSH testing. At that time, patient L was taking thyroid from an independent source. After petitioner stipulated to the ISO and after the patient’s TSH level was recorded within the normal range, petitioner prescribed thyroid for patient L for the first time, and he continued treating L with thyroid when he saw L in the following months. Based on those findings, the board concluded that petitioner’s treatment of patient L violated the ISO’s requirement, in paragraph 3.1, that petition not “prescribe, or direct any patient to take thyroid unless patient TSH levels exceed 10 uIU/mL[.]”
On review, petitioner argues that paragraph 3.1 does not apply; he claims that paragraph 3.2 applies instead and that he complied with its requirements to have “any patient taking thyroid from a non-prescription source * * * [178]*178undergo thyroid blood tests on a regular basis (at least every 6 months) ***.” Petitioner contends that “[p]aragraph 3.1 of the ISO refers to the decision to initiate treating patients on thyroid, whereas 3.2 involves treatment of patients who are already on thyroid (thus, requiring blood tests every six months, which [petitioner] complied with).” (Underscoring in original.) We disagree. Paragraph 3.1 imposes restrictions in prescribing thyroid to “any patient”; paragraph 3.2 imposes restrictions on treatment for patients taking thyroid from a nonprescription source. And after the ISO went in effect, petitioner for the first time prescribed thyroid to patient L, even though the only reported TSH testing for patient L did not “exceed 10 uIU/mL[.]” Therefore, the board properly determined that paragraph 3.1 applied and petitioner violated that paragraph in treating patient L.
As to patient I, the board concluded that petitioner’s treatment violated the ISO’s requirement to “comply with the guidelines recommended in the article ‘Risks of Testosterone Replacement Therapy and Recommendations for Monitoring,’ published in the New England Journal of Medicine [.]” Relying on testimony from Cook, the board concluded that the New England Journal of Medicine (NE JM) article presupposed a proper diagnosis of hypogonadism based on low levels of serum testosterone, but patient I’s testosterone levels (taken many years before, when patient I first saw petitioner) were normal and would not qualify for the diagnosis for hypogonadism.
Petitioner contends that the board’s understanding of the NE JM article — and, in turn, its interpretation of the ISO — was wrong because the article “does not provide an ‘absolute number in the guidelines’ for purposes of determining when to prescribe testosterone hormones.” We agree. Although the NEJM article references “low levels of serum testosterone” as one indicator for hypogonadism, it does not define those “low levels” for an appropriate diagnosis.4 It [179]*179was Cook’s medical opinion, not the NEJM guidelines, that established that patient I’s “testosterone levels were within normal limits.” Because nothing in the NEJM article or the ISO sets forth the testosterone levels for a “proper diagnosis,” the board erred in concluding that petitioner’s treatment of patient I violated the ISO.
We turn to petitioner’s broader claim in his fourth assignment of error — covering the remaining patients at issue on review — that the board erred in concluding that his treatment did not constitute “alternative medical treatment” as defined in ORS 677.190(l)(b)(A). Petitioner’s argument proceeds in two steps. He first contends that his treatment of all patients met the three statutory criteria for “alternative medical treatment” under ORS 677.190(l)(b)(A):
“(A) ‘Alternative medical treatment’ means:
“(i) A treatment that the treating physician, based on the physician’s professional experience, has an objective basis to believe has a reasonable probability for effectiveness in its intended use even if the treatment is outside recognized scientific guidelines, is unproven, is no longer used as a generally recognized or standard treatment or lacks the approval of the United States Food and Drug Administration;
“(ii) A treatment that is supported for specific usages or outcomes by at least one other physician licensed by the Oregon Medical Board; and
“(iii) A treatment that poses no greater risk to a patient than the generally recognized or standard treatment.”
The board erred, petitioner argues, by ignoring the statutory definition and concluding that his treatment was not “alternative medical treatment” simply because it was “conventional treatments in a non-conventional manner.” According to petitioner, he was practicing “alternative medical treatment” when he prescribed or recommended hormones to patients based primarily on the clinical presentation (i.e., symptoms that could be associated with a hormone deficiency), even if no hormone testing was conducted or if testing showed that a patient’s hormone levels were within a recognized normal range. Next, petitioner claims that his conduct cannot be found to violate the standard of care or [180]*180constitute unprofessional conduct5 because “[treatment which falls within the statutory definition of [alternative medical treatment] is exempt from the [board’s] general ‘standard of care’ interpretation.” (Underscoring omitted.) He contends that that is what ORS 677.190(l)(b) means in providing that “the use of an alternative medical treatment shall not by itself constitute unprofessional conduct.” (Emphasis added.) Petitioner claims that a doctor using alternative medical treatment is only held to a “general medical standard of care,” which, for example, would be violated by using unsterilized needles.
In response, the board contends that petitioner’s hormone replacement treatment was not alternative medical treatment under ORS 677.190(l)(b)(A) because petitioner did not have an objective basis for believing that the treatment would be effective and the treatment “did pose greater risk to his patients than the generally recognized use of [hormone] supplements.” Further, the board argues that “[n]othing in ORS 677.190(l)(b) prohibits the board from regulating alternative] medical treatment.” According to the board, because ORS 677.190(l)(b) provides only that “the use of an alternative medical treatment shall not by itself constitute unprofessional conduct,” that statute “has no bearing on whether [petitioner] met the applicable standard of medical care.”
We need go no further than the first step of petitioner’s argument. The board specifically concluded that, even if the first two criteria of ORS 677.190(l)(b)(A) were satisfied, petitioner’s “treatment posed a greater risk of harm to his patients than the standard or recognized treatment, contrary to the provisions of ORS 677.190(l)(b)(A)(iii).” On review, petitioner challenges the board’s finding that petitioner’s hormone treatment, without establishing a deficiency through hormone testing, exposed patients to certain risks resulting from receiving too much hormone. Our review is limited to determining whether “the record, viewed as a whole, would permit a reasonable person to make that finding.” ORS 183.482(8)(c); see Spray v. Bd. of [181]*181Medical Examiners, 50 Or App 311, 318-19, 324, 624 P2d 125, modified on recons, 51 Or App 973, 627 P2d 25, rev den, 291 Or 117 (1981) (stating that expert testimony determines the standard of conduct for unprofessional conduct and that, where expert witnesses disagree as to adequacy or appropriateness of treatment, the court reviews the board’s findings based on that testimony for substantial evidence). When viewing the record as a whole, we conclude that substantial evidence — in this case, expert testimony — supports the board’s finding that petitioner’s treatment posed a greater risk of harm than the generally recognized treatment.
As to petitioner’s thyroid treatment, the board found that, because patients who exhibited symptoms consistent with a thyroid deficiency might not have a deficiency (;i.e., the symptoms were the result of some other condition), providing thyroid without establishing a medical deficiency exposed those patients to possible “accelerated bone density loss, or osteopenia, cardiac complications including cardiac arrhythmia, and developing hyperthyroidism, which may increase the risk for hypertension.” That finding is supported by substantial evidence in the record: both Nedrow and Cook testified as to the risk of harm generally and on a patient-by-patient basis.6
Substantial evidence also supports the board’s finding that petitioner’s testosterone treatment posed a greater risk than the generally recognized use of that hormone in treatment. In testimony describing petitioner’s treatment of F with testosterone, for example, (1) Cook testified that he didn’t “think [F] needed testosterone” and agreed that “giving a patient too much testosterone [could] overstimulate prostate growth” and “lead to urinary, bladder obstruction”; [182]*182(2) Nedrow testified that, with respect to patient F, petitioner did not meet the standard of care because, as with “elite athletes, we just don’t think it’s worth the risks to give anabolic steroids to people that do not have a medical deficiency,” and
(3) patient F’s primary care physician testified that he was concerned about the risk that excessive testosterone would contribute to patient F’s high blood pressure, which had not been well controlled over several years of treatment.7 As to patient G, who was diagnosed with prostate cancer, the board, acknowledging contrary expert testimony, found that “[t]estosterone replacement therapy is linked to exposing underlying cancers and to accelerating tumor growth” and that “[e]vidence supporting the recommendation to not treat men with documented testosterone deficiency and prostate cancer with testosterone replacement therapy has been found overwhelming and valid.”8
In light of the board’s findings as to the risk of harm from petitioner’s thyroid and testosterone treatment, we conclude that the board did not err in concluding that petitioner’s treatment put his patients at a risk of harm greater than the standard treatment and his treatment therefore did not qualify as “alternative medical treatment” under ORS 677.190(l)(b)(A). Accordingly, we reject petitioner’s claim that the board erred in concluding that he breached the standard of care because “[treatment which falls within the statutory definition of [alternative medical treatment] is exempt from the [board’s] general ‘standard of care’ interpretation.”9 (Underscoring omitted.)
[183]*183One last issue remains for our consideration. Apart from his argument regarding alternative medical treatment, petitioner presents patient-specific arguments that the board erred in concluding that his treatment of patients A through E with thyroid breached the standard of care or violated the 2004 order.10 We reject those arguments with respect to patients B, D, and E. Although petitioner points out that those three patients were taking thyroid when they first began treatment with petitioner, the board correctly concluded that petitioner violated the 2004 order and the standard of care by diagnosing hypothyroidism and continuing to monitor that condition without conducting TSH testing, failing to establish a baseline thyroid level (i.e., continuing treatment with thyroid without testing to see what the patient’s TSH levels were without thyroid treatment), or failing to regularly test thyroid levels on an annual basis.
As for patients A and C, petitioner argues that they “received no medical treatment from [petitioner], as they each saw [him] on just one occasion, and elected not to follow-up with treatment.” According to petitioner, it follows that petitioner did not violate the 2004 order’s requirements to “use thyroid function blood tests * * * in conjunction with the history and physical findings in making the decision whether to use thyroid medication” and to “not use thyroid medication in treatment unless the blood tests find a TSH level greater than the normal range and a free T4 below the normal range on a test.” (Emphases added.)
The board did not err in concluding that petitioner violated the 2004 order when treating patient A. Although patient A’s lab results (reported earlier by her primary care physician) showed normal thyroid levels, petitioner told patient A that she had problems with her thyroid, that taking thyroid would help, and how to get thyroid online from [184]*184Mexico.11 We agree with the board that petitioner violated the 2004 order because he “ma[de] the decision” that thyroid treatment was appropriate based on his physical examination alone. Further, when petitioner communicated that opinion to the patient and instructed patient A how to get thyroid from an outside source, petitioner “use[d] thyroid medication in treatment” within the meaning of the 2004 order.12
We agree with petitioner, however, that he did not violate the 2004 order when he saw patient C at a single visit. Petitioner told patient C that, based on her family history and symptoms, she probably needed thyroid; he recorded a diagnosis of probable hypothyroid type 2 in his notes; and he recommended TSH and T4 testing. Petitioner’s qualified diagnosis, without any recommendation that patient C take thyroid from any source, did not show that petitioner had “ma[de] the decision” to use thyroid in treating patient C. Nor did that qualified diagnosis give rise to the “use” of thyroid in treatment. The board erred in concluding that petitioner’s treatment of patient C violated the 2004 order.13
[185]*185In sum, we affirm the board’s order in all respects, except that we conclude that petitioner’s treatment of patient I did not violate the ISO and his interaction with patient C did not violate the 2004 order. Because the board’s order does not state that the sanctions imposed would be warranted without those violations of prior board orders, we reverse and remand for the board to consider the appropriate sanctions.
Reversed and remanded for reconsideration.