HASELTON, P.J.

Reversed and remanded for reconsideration.

HASELTON, P.J.

Petitioner, a nurse practitioner, seeks judicial review of a final order of the Board of Nursing (board) approving an order of emergency suspension and suspending petitioner from the practice of nursing for at least two years. Petitioner argues, principally, that the board erred in determining that certain findings of historical fact made by the administrative law judge (ALJ) were not supported by a preponderance of the evidence in the record and in modifying those findings pursuant to ORS 183.650(3). As amplified below, on de novo review of the record pursuant to ORS 183.650(4), we determine that the board did not err in modifying the ALJ's findings of historical fact pertaining to petitioner's conduct involving patients SM, SC, ES, and DB. However, we also determine that the board erred in modifying certain findings of historical fact pertaining to petitioner's alleged sexual misconduct and financial misconduct involving patient WJ. Accordingly, we remand to the board for reconsideration and entry of an order consistent with this opinion. See ORS 183.650(4).

A detailed discussion of the allegations against petitioner and of the evidence presented at the contested case hearing would be of no value to the bench, bar, or parties. Except as supplemented below, it suffices to say that petitioner was a family nurse practitioner with offices in two south-central Oregon communities. In the course of his practice, petitioner was authorized to conduct physical examinations of patients, including breast and pelvic examinations of female patients. Petitioner had been licensed as a registered nurse since 1971 and had not previously been subject to disciplinary action by the board.

On June 20, 2002, the board issued an order of emergency suspension and a notice of proposed indefinite suspension against petitioner after investigating complaints involving a number of petitioner's female patients. For the most part, those complaints concerned petitioner's alleged actions in conducting physical examinations. The board later amended both the initial order and notice and ultimately issued its third amended order of emergency suspension and second amended notice of proposed indefinite suspension on August 7, 2002. In the third amended order of emergency suspension, the board recited that it had received and investigated "independent, credible complaints" that petitioner had engaged in sexual contact with five female patients, WJ, SM, SC, ES, and DB, who were "vulnerable to exploitation due to age [and] mental and emotional health issues"; that there was "a high probability of recurrent acts of sexual misconduct and boundary violations" and, consequently, because petitioner posed a serious danger to public health and safety, his license was immediately suspended under ORS 183.430(2). The board's second amended notice of proposed indefinite suspension alleged the same conduct, with an additional allegation that petitioner had inappropriately accepted $7,000 from WJ. The board alleged that petitioner's conduct violated ORS 678.111(1)(f) and (2), and OAR 851-045-0015(1)(l) and (m), and (2)(a), (c), and (g).

A contested case hearing was held before the ALJ in late July and early August 2002. During that hearing, more than a dozen witnesses testified, including petitioner, members of petitioner's office staff, the patients SM, SC, and ES, WJ's granddaughter, and various expert witnesses. The ALJ also received an affidavit from WJ and a recorded statement by DB, as well as other exhibits, including medical records. The ALJ issued a 19-page proposed order, which included 35 "findings of fact," that ultimately determined that the board's allegations against petitioner were unfounded.

The ALJ was not authorized to issue a final order. See ORS 183.464. Accordingly, the board subsequently issued an amended proposed order. After petitioner filed extensive exceptions to that proposed order and the board rejected those exceptions, the board, in February 2003, issued its 51-page final order, which is the object of our review.

The board's final order differs dramatically from the ALJ's proposed order in several respects. First, the board added an extensive, detailed assessment of petitioner's credibility and of the credibility of various other witnesses, including one of petitioner's assistants, Renfro, and the patients SC, ES, and SM. Unlike the ALJ, who had implicitly found petitioner and Renfro to be credible by accepting their versions of disputed facts, the board flatly rejected petitioner's and Renfro's credibility on a variety of non-demeanor-based grounds.

Conversely, unlike the ALJ, who had determined that the various patients had either misrepresented or misperceived petitioner's actions, the board explicitly determined that SC, ES, and SM were credible and that the hearsay-based allegations of misconduct towards WJ and DB were reliable. The board's determinations with respect to the credibility of the patient witnesses' testimony and the reliability of related hearsay were based on a variety of factors, including cross-corroboration from similarities among the patients' accounts (all patients), lack of motive to fabricate (SC, ES), reputation in the community (SC), consistency with prior statements and contemporaneous reports (SC, ES, SM), and the inapplicability of the post-traumatic stress disorder-related "mistaken cue" dynamic that the ALJ had relied on in discounting the testimony of SC and ES. Second, given its determination that petitioner and Renfro were not credible, the board modified the ALJ's proposed findings of fact by deleting those findings that were based on their uncorroborated testimony. The board also deleted other proposed findings by the ALJ, deeming them to be immaterial. Conversely, given its determination that the patient witnesses were credible, the board modified the ALJ's proposed findings by adding material conforming to those witnesses' accounts. Further, the board modified the findings regarding petitioner's alleged misconduct against the patient WJ based on its determination that third-party testimony and hearsay pertaining to those allegations were reliable.

Third, based on its endorsement of the expert testimony of Dr. Steven Mussack and its discounting of the testimony of petitioner's expert Dr. Faulder Colby, see 197 Or App at 522 n 4, the board adopted additional or supplemental findings of fact relating to the dynamics of sexual misconduct by medical professionals.

Fourth, based on its findings of fact, which incorporated the foregoing credibility determinations, deletions, additions, and supplementations, the board concluded that petitioner had engaged in conduct derogatory to the practice of nursing in 11 particulars and had not engaged in such conduct in three other particulars. The board concluded:

"16. Licensee's continued exercise of his license privileges posed a serious danger to the public health or safety based on the serious misconduct committed as to Patients WJ, SM, and SC standing alone or together with Patients ES and DB, and the strong likelihood that such misconduct would continue, such that his nursing license and certificate were correctly immediately suspended, pursuant to ORS 183.430(2).

"17. Licensee's nursing license and certificate should be suspended for not less than two (2) years and, before being reinstated, Licensee is required to undergo and complete appropriate evaluation, treatment, and education, for having committed the foregoing conduct derogatory to the practice of nursing."

The board also confirmed the correctness of its earlier emergency suspension of petitioner's license.

On judicial review, petitioner raises a battery of challenges to the board's findings, reasoning, and disposition. We reject most of those challenges without discussion but write to address one matter: Under ORS 183.650(4), did the board err in modifying the ALJ's findings of historical fact with respect to the order of indefinite suspension?

In 1999, the legislature enacted House Bill (HB) 2525, which transformed Oregon administrative practice, and related judicial review, with respect to most contested case proceedings. The impetus for that legislation was that administrative agencies acted as both the "prosecutor" and the "adjudicator" in contested cases, giving rise to concern about inherent conflict and perceptions of bias. See, e.g., Testimony, House Judiciary Committee on Civil Law, HB 2525, Apr 26, 1999, Ex Q 1-2 (written statement of John DiLorenzo, Jr.). Accordingly, the legislature (1) created a central pool of independent "hearing officers," in lieu of "in-house" hearing officers who worked exclusively for a particular agency, and (2) directed that, with the exception of specifically identified agencies, see ORS 183.635(2), (3), all agencies must use hearing officers from the Office of Administrative Hearings to conduct contested case hearings. See ORS 183.605, ORS 183.635(1). The Board of Nursing was not among the agencies excepted from that legislation. See ORS 183.635(2).

Two other features were crucial to the legislative design. First, the legislature redefined the function of the administrative agencies in reviewing and modifying a proposed form of order issued by an administrative law judge. ORS 183.650 provides, in part:

"(1) In any contested case hearing conducted by an administrative law judge assigned from the Office of Administrative Hearings, the administrative law judge shall prepare and serve on the agency and all parties to the hearing a form of order, including recommended findings of fact and conclusions of law. * * *

"(2) If the administrative law judge assigned from the office will not enter the final order in a contested case proceeding, and the agency modifies the form of order issued by the administrative law judge in any substantial manner, the agency must identify the modifications and provide an explanation to the parties to the hearing as to why the agency made the modifications.

"(3) An agency conducting a contested case hearing may modify a finding of historical fact made by the administrative law judge assigned from the Office of Administrative Hearings only if the agency determines that the finding of historical fact made by the administrative law judge is not supported by a preponderance of the evidence in the record. For the purposes of this section, an administrative law judge makes a finding of historical fact if the administrative law judge determines that an event did or did not occur in the past or that a circumstance or status did or did not exist either before the hearing or at the time of the hearing."

(Emphasis added.)

Second, the legislature, in ORS 183.650(4), also redefined this court's standard of review with respect to "findings of historical fact" when the agency has modified such a finding by the ALJ. Under prior and long-standing law, our review of an agency's findings, including findings of historical fact, had been limited to determining whether those findings were supported by substantial evidence in the record viewed as a whole. See, e.g., Kniss v. PERB., 184 Or App 47, 52, 55 P3d 526 (2002). However, under ORS 183.650(4),

"[i]f a party seeks judicial review of an agency's modification of a finding of historical fact under subsection (3) of this section, the court shall make an independent finding of the fact in dispute by conducting a review de novo of the record viewed as a whole. If the court decides that the agency erred in modifying the finding of historical fact made by the administrative law judge, the court shall remand the matter to the agency for entry of an order consistent with the court's judgment."

Thus, pursuant to subsection (4), in contested case proceedings in which an ALJ from the central Office of Administrative Hearings has rendered "findings of historical fact," and the agency, in its final order, has modified those findings, our standard of review is de novo — and not for "substantial evidence." In engaging in de novo review under subsection (4), we apply a "preponderance of evidence" standard. Becklin v. Board of Examiners for Engineering, 195 Or App 186, 204-05, 97 P3d 1216 (2004), rev den, 338 Or 16 (2005).

HB 2525 fundamentally altered the roles of the agencies and courts with respect to most contested case proceedings. And yet, we have engaged in de novo review pursuant to ORS 183.650(4) only once before. See Becklin, 195 Or App at 204-06. Consequently, many questions regarding the statute's operation remain unresolved.

For example, what exactly is a "finding of historical fact"? As noted, ORS 183.650(3) defines that term as a

"determin[ation] that an event did or did not occur in the past or that a circumstance or status did or did not exist either before the hearing or at the time of the hearing."

Some matters clearly fall within the contours of that definition. An ALJ's determination that "the light was green" is a "finding of historical fact" — and the agency's subsequent modification of that finding to read, "the light was red" would trigger our de novo review under ORS 183.650(4).

The proper characterization and treatment of other matters may be less simple. Consider a case in which an ALJ finds that an event occurred and does so based on explicit "findings" regarding a witness's credibility, and then the agency renders a finding that the event did not occur — and, in so doing, expressly finds that the witness on whom the ALJ relied is not credible. Is the agency's modification of the ALJ's credibility findings a modification of "a finding of historical fact" subject to our de novo review under ORS 183.650(4)? A similar question may arise when an ALJ's finding of what occurred, and the agency's contrary finding, are each predicated on an assessment of the persuasiveness of expert testimony relating to the likelihood of misperception and misreporting of an occurrence. Again, is the agency's modification of the ALJ's findings regarding such expert testimony a matter subject to de novo review under ORS 183.650(4)?

A second unsettled area concerns the proper limits of our de novo review. If the ALJ found that "the light was green," and the agency that "the light was red," is our review under ORS 183.650(4) limited to "green" or "red" — or, if we so determine by a preponderance of the evidence, can we find, de novo, that "the light was yellow"? Conversely, if we believe, after engaging in de novo review, that all parties' evidence is unreliable, can we state, "We are not persuaded that the light was red (as found by the agency) or that it was green (as found by the ALJ) — and, thus, both findings of historical fact were in error"?

Both of the foregoing matters are, to some extent, implicated here. However, before addressing their substance as pertinent, we must give a clear procedural warning to all future petitioners who seek to invoke our de novo review pursuant to ORS 183.650(4): Any petitioner who asks this court to engage in de novo review under ORS 183.650(4) must specifically identify each challenged modification of a finding of historical fact and explain why that modification was erroneous as unsupported by a preponderance of the evidence.

Why? Final agency orders are often (as here) extensive documents, consisting of dozens of findings and conclusions, with interrelated analyses. As noted, not all findings are "findings of historical fact" within the meaning of ORS 183.650(4). Further, as noted, not all "findings of historical fact" in an agency's final order represent "modifications" of findings of historical fact by the ALJ. See Becklin, 195 Or App at 206. Thus, an omnibus invocation of ORS 183.650(4) is an invitation to appellate chaos.

We are entitled to reject such an invitation. See ORAP 5.45(3), (4)(c). We have repeatedly emphasized that it is not this court's obligation "to search the record to find the error." ORAP 5.45(4)(c); see, e.g., Wahlgren v. DMV, 196 Or App 452, 457, 102 P3d 761 (2004); Resources Northwest, Inc. v. Crothers, 153 Or App 24, 26-27, 955 P2d 763 (1998). Further:

"[I]t is not this court's function to speculate as to what a party's argument might be. Nor is it our proper function to make or develop a party's argument when that party has not endeavored to do so itself."

Beall Transport Equipment Co. v. Southern Pacific, 186 Or App 696, 701 n 2, 64 P3d 1193, adh'd to on recons, 187 Or App 472, 68 P3d 259 (2003).

Thus, to the extent that a future petitioner merely raises an omnibus or otherwise ill-focused challenge under ORS 183.650(4) to the correctness of the findings of fact in an agency's final order, we will reject that challenge as unreviewable. Rather, it is incumbent upon the petitioner to (1) separately identify each challenged finding of historical fact in the agency's final order; (2) identify the finding of historical fact in the ALJ's proposed order that the agency's finding "modified"; and (3) explain why, with respect to each such "modified" finding, the agency's finding is contrary to the preponderance of the evidence.

Here, petitioner's appellate brief does not comport with those requirements. Instead, petitioner's assignments of error embody, for the most part, broad-based challenges to the manner in which the board determined the facts and modified the ALJ's proposed findings generally, coupled with a description of "[t]he most glaring examples of historical facts supported by the preponderance of evidence in the record that the board determined to modify in spite of the law[,]" including determinations of credibility. Nevertheless, we fully appreciate that ORS 183.650(4) presents novel challenges for practitioners, as well as judges, and that we have not previously addressed the procedural interplay between ORAP 5.45 and ORS 183.650(4). Accordingly, as a matter of fairness, we apply the procedural requirements announced here prospectively only. See generally Peiffer v. Hoyt, 186 Or App 485, 497-98, 63 P3d 1273, rev allowed, 336 Or 16 (2003) (describing circumstances in which Court of Appeals has applied procedural rule prospectively only).

We return, then, to the merits. Before actually undertaking de novo review pursuant to ORS 183.650(4), however, we must determine the proper scope of such review. That inquiry, in turn, requires us to revisit and resolve several of the questions posited above concerning the operation of ORS 183.650(4).

We start with the content of the term "finding of historical fact." As noted, consistently with the definition in ORS 183.650(3), some matters are obviously matters of "historical fact," e.g., whether a company used a particular chemical in its operations, whether an officer gave certain warnings to a suspect — or, here, whether petitioner actually engaged in specific acts or made specific comments to his patients. In this case, however, the board's determinations that petitioner had, in fact, engaged in actionable misconduct were inextricably intertwined with its assessment of two other matters: (1) the witnesses' credibility (and, particularly, the credibility of petitioner, Renfro, and the patient witnesses); and (2) the persuasiveness of expert testimony regarding the dynamics of sexual misconduct by medical professionals and the potential for misperception of conduct by patients with histories of abuse and related stress disorders. Here, neither of those matters is explicitly a "matter of historical fact" in the sense of whether "was the light green or was it red?" And yet — in what was, at its core, a "swearing match" dispute — the ALJ's and the board's assessments of each drove their respective determinations as to what did (or did not) occur between petitioner and his patients.

The upshot of the foregoing is that, regardless of whether witness credibility or the persuasiveness of expert testimony is itself a matter of "historical fact" per se, either may be so essential to our de novo determination of a matter of "true" historical fact that we cannot find the latter without making our own assessment of the former. That is the case here. Accordingly, as amplified below, our de novo determinations as to whether petitioner did, in fact, engage in the alleged acts that are the bases of the board's final order of indefinite suspension are predicated on our own, independent assessments of the witnesses' credibility and the persuasiveness of the expert opinions of Mussack and Colby.

A second potential area of uncertainty that we identified above pertained to whether, on de novo review, we are limited to choosing between the historical facts as found by the ALJ ("the light was green") or by the agency ("the light was red") — or whether we are authorized, indeed obligated, to make our own determinations of the disputed matter ( viz., "What color was the light?"). The statute's text is unambiguous and conclusive: "[T]he court shall make an independent finding of the fact in dispute by conducting a review de novo of the record viewed as a whole." ORS 183.650(4) (bold-face added). Bluntly: The statute does not say that we are to determine, as between the ALJ's and the agency's conflicting findings, which comes closer to the historical "truth." Rather, we are to find the disputed fact(s) "independent[ly]." Thus, if we determine on de novo review that the preponderance of the evidence in the record establishes that "the light was yellow," we must so find.

By the same token, in our role as de novo factfinder, we are — and must be — free to determine that the evidence in the record is so unreliable that we cannot determine "the fact in dispute" by a preponderance of the evidence. Assume a case in which the determination of the stoplight's color rests solely on choosing between witness A, who testified that the light was green but whose credibility has been thoroughly impeached, and witness B, who testified that the light was red but who is delusional and has given contradictory accounts. In that circumstance, even if the ALJ, relying on A's testimony, had found that "the light was green," and the agency, relying on B's testimony, had found that "the light was red," we might well determine that the evidence was insufficient to permit any finding of the light's color.

Still, what is the proper disposition under ORS 183.650(4) if we either (1) affirmatively find the fact in dispute differently from either the ALJ or the agency; or (2) determine that we cannot find the fact in dispute by a preponderance of the evidence? The statute provides:

"If the court decides that the agency erred in modifying the finding of historical fact made by the administrative law judge, the court shall remand the matter to the agency for entry of an order consistent with the court's judgment."

Returning again to our stoplight illustration, if the ALJ found that the light was green, the agency found that the light was red, and we find that the light was yellow, the agency would not, in one sense, have "erred in modifying the finding of historical fact" made by the ALJ — after all, the ALJ's finding that the light was green was, in fact, inaccurate. However, in a more precise sense, the agency "erred in modifying" the ALJ's finding because, in doing so, the agency itself rendered an inaccurate finding. The same would be true of the second, "insufficient evidence" scenario.

Given the statute's design, it is apparent that the legislature intended that we remand to the agency pursuant to ORS 183.650(4) in such circumstances. That is, we are to remand not only when we determine on de novo review that the ALJ was "right" and the agency was "wrong" about a disputed matter of historical fact. Rather, we are also to remand when we find that both the ALJ and the agency were "wrong" — either because we affirmatively find the fact in dispute differently than either or because we determine that the evidence in the record does not preponderate on the disputed matter.

With our de novo review function under ORS 183.650(4) so understood and defined, we return to this case. Upon de novo review of the record as a whole, we determine that the board did not err in modifying the ALJ's findings of historical fact pertaining to petitioner's alleged misconduct towards his patients, SM, SC, ES, and DB. With respect to the substance of those modifications, "we independently find, by a preponderance of the evidence, the same facts as the board found." Becklin, 195 Or App at 206. Our findings in that regard, like the board's, are grounded primarily in our assessment of the credibility of those patients, as opposed to that of petitioner and Renfro, and the persuasiveness of Mussack's expert testimony as opposed to Colby's.

We note particularly, but without limitation, several salient features bearing on the patient witnesses' credibility: (1) SC was an especially impressive witness, who had absolutely no reason to fabricate her allegations against petitioner and, given her standing in her community had, (as the board put it), "much to lose by testifying against [petitioner]." (2) There were significant "cross-collaborative" common elements among the accounts of SM, SC, ES, and DB. (3) SM, SC, and ES told third parties, including counselors, about petitioner's alleged conduct before reporting that conduct to the board. (4) Mussack's testimony, which we accept, persuasively explains why female victims of sexual abuse by medical professionals may delay reporting such abuse or may continue seeking treatment from the professional even after reporting the misconduct to others. (5) Like the board, we reject petitioner's assertion, based on Colby's testimony, that SC's and ES's accounts were not credible because they were the product of misperception or misinterpretation arising from those patients' post-traumatic stress disorders associated with prior abuse.

Conversely, our bases for finding petitioner and Renfro not to be credible on matters of disputed fact pertaining to the alleged misconduct parallel those that the board described in detail in its final order. Those reasons include material inconsistencies in petitioner's testimony, as well as petitioner's and Renfro's motivations to minimize, overstate, or fabricate.

That does not, however, conclude our review. The board also modified the ALJ's findings of historical fact regarding petitioner's alleged misconduct involving the fifth patient, WJ. Whereas the ALJ had found that petitioner had not engaged in either sexual misconduct or financial misconduct involving WJ, the board found that petitioner had engaged in both types of misconduct. On de novo review, we find that the board's findings regarding sexual misconduct towards WJ and some aspects of the board's findings regarding financial misconduct were not supported by a preponderance of the evidence.

With respect to the allegations of sexual misconduct involving WJ, this case is similar to an illustration posited above: This is, in essence, a "swearing match" in which there is no reliable evidence on either side. Because there is no reliable evidence, we cannot find by a preponderance of the evidence either that the alleged misconduct occurred or that it did not occur.

WJ did not testify at the hearing. Instead, the board's proof of the alleged sexual misconduct consisted of testimony by WJ's granddaughter, recounting statements that WJ had made to her, and chartnotes in which a physician had memorialized similar statements by WJ. We find, as did the board, that the granddaughter's testimony and the chartnotes both accurately recounted WJ's statements. However, unlike the board, we do not accept WJ's statements as reliable.

That is so for three reasons. First, at the time of the alleged abuse and her ensuing statements, WJ was being treated, or had recently been treated, for bipolar disorder, "early Alzheimer's," and presenile dementia with visual, auditory, and olfactory hallucinations. Second, at the time of the hearing, WJ executed an affidavit, which petitioner's counsel submitted, in which she denied that petitioner had engaged in any mistreatment. Third, the particular conduct and the circumstances of the abuse that WJ described to her granddaughter and her physician differed substantially from the other patients' accounts. Given WJ's highly questionable competence, the inconsistencies in her statements, and the lack of meaningful cross-corroboration of the accuracy of her allegations of abuse, we cannot find by a preponderance of the evidence that petitioner engaged in sexual misconduct involving WJ. Thus, the board erred in modifying the ALJ's findings by rendering findings of historical fact that such abuse had occurred.

We emphasize, however, that our determination does not sustain the accuracy of the ALJ's obverse findings of historical fact, which, as an affirmative matter, adopted petitioner's account of his interactions with WJ. That is, we do not find by a preponderance of the evidence that petitioner did not engage in sexual misconduct involving WJ. Given that we have rejected petitioner's denials of sexual misconduct involving four other women patients, we do not give credence to his denial of such misconduct towards WJ. In sum, we cannot find by a preponderance of the evidence either that petitioner engaged in sexual misconduct involving WJ or that he did not.

We reach the same determination with respect to certain disputed matters of historical fact pertaining to petitioner's alleged financial misconduct involving WJ. It is undisputed that petitioner accepted substantial cash payments from WJ that were not paid as compensation for services. However, the parties disagreed regarding the circumstances of those payments, and the board substantially modified the ALJ's findings of historical fact regarding those circumstances.

Specifically, the ALJ found that WJ, unilaterally and without solicitation, gave petitioner (and, in one instance, petitioner's wife) those monies as "gifts"; that petitioner initially refused to accept each of those gifts and returned or attempted to return them; and that petitioner ultimately accepted the monies only after consulting with counsel and after WJ consistently rejected his efforts to return the funds. Conversely, the board issued a modified finding that petitioner had asked WJ for the funds to help pay for equipment and his clinic's rent. The board also deleted the ALJ's findings describing petitioner's initial resistance and ultimate reluctant acceptance of the payments. Based on those modifications, the board concluded that petitioner had improperly "exploited" his professional relationship with WJ "by gaining * * * [an] item of value from the client either for personal gain or sale, beyond the compensation for nursing services." OAR 851-045-0015(2)(g).

Given those modifications, the facts in dispute on de novo review are (a) whether petitioner solicited some or all of the payments and (b) whether, even if petitioner did not ask for the payments, he attempted to return the "gifts" and accepted them only after WJ rejected those efforts. For the same reasons that we cannot render findings of fact regarding the alleged sexual misconduct involving WJ, we cannot do so regarding whether petitioner solicited, or WJ unilaterally gave, some or all of the cash payments. There is no reliable evidence in the record on that matter and, thus, no basis for a finding of fact by a preponderance of the evidence. Again, WJ's competence is extremely questionable, and she gave inconsistent statements; again, we do not regard petitioner's uncorroborated testimony on disputed matters regarding alleged misconduct as credible. Accordingly, the board's modified finding of historical fact that petitioner solicited funds from WJ was erroneous as unsupported by a preponderance of the evidence.

Finally, we determine that the board did not err in deleting the ALJ's findings that petitioner had consistently refused to accept cash "gifts" from WJ and had accepted such payments only after WJ had rebuffed efforts to return them. The ALJ's findings in that regard were predicated solely on petitioner's uncorroborated testimony. Because we, like the board, do not accept petitioner's testimony on those matters as credible, we agree that those findings were properly deleted as unsupported by a preponderance of the evidence.

In sum, after engaging in de novo review pursuant to ORS 183.650(4), we have reached the following determinations:

(1) The board did not err in modifying the ALJ's findings of historical fact regarding sexual misconduct towards patients SM, SC, ES, and DB. Based on our independent consideration of the record viewed as a whole, we find that the board's modified findings comport with the preponderant evidence.

(2) The board did err in modifying the ALJ's findings of historical fact regarding petitioner's alleged sexual misconduct towards patient WJ. There is no reliable evidence — and, thus, there cannot be preponderant evidence — that the alleged misconduct occurred, or did not occur.

(3) For the same reasons described in (2), the board erred in modifying the ALJ's findings of historical fact so as to affirmatively find that petitioner solicited cash payments from WJ. There is no reliable evidence either that petitioner asked for such payments or that WJ unilaterally made those payments as "gifts" to petitioner.

(4) The board properly deleted the ALJ's findings that petitioner attempted to refuse and to return the cash payments to WJ. We do not accept petitioner's testimony regarding those matters as credible.

Reversed and remanded for reconsideration.