Texas State Board of Medical Examiners and Donald W. Patrick, M.D., J.D., Executive Director of the Texas State Board of Medical Examiners v. Jack Dunn, III

TEXAS COURT OF APPEALS, THIRD DISTRICT, AT AUSTIN NO. 03-03-00180-CV Texas State Board of Medical Examiners and Donald W. Patrick, M.D., J.D., Executive Director of the Texas State Board of Medical Examiners, Appellants v. Jack Dunn, III, Appellee FROM THE DISTRICT COURT OF TRAVIS COUNTY, 345TH JUDICIAL DISTRICT NO. GN104061, HONORABLE SUZANNE COVINGTON, JUDGE PRESIDING M E M O R A N D U M O P I N I O N In this case, we address the circumstances under which the Texas State Board of Medical Examiners may reject an Administrative Law Judge's findings of fact and substitute its own. We hold that the Board, which does not have unlimited discretion to change an ALJ's findings of fact and conclusions of law, did not establish a reasonable evidentiary basis for rejecting the ALJ's findings and conclusions. For the following reasons, we affirm the judgment of the district court. Appellee Jack Dunn, III, was a practicing anesthesiologist in July 1995, when he administered an epidural to M.J., a woman who went into cardiac arrest and died during childbirth. In September 1997, the Board and Dunn entered into an agreed order under which his license was suspended until he showed that he was able to safely practice medicine (the "Agreed Order"). According to the Agreed Order, at the time of M.J.'s death, Dunn was abusing prescription drugs. The Agreed Order also contained findings of fact that during 1995 and 1996, other doctors and nurses raised concerns about Dunn's job performance and competency; he was removed as a member of a physician group; his privileges at a Lubbock hospital were suspended; he improperly prescribed prescription drugs to a female patient with whom he had a romantic relationship; and he settled for one million dollars a malpractice suit brought as a result of M.J.'s death. About one year after his suspension, Dunn began to seek to have his suspension lifted. In January 1999, an informal settlement conference ("ISC") was held before a panel including one or more Board members. The ISC panel recommended that Dunn's suspension be lifted subject to conditions that included practice restrictions and monitoring requirements. When the recommendation was presented to the full Board, it was rejected. Another ISC was held in October 1999, the panel again recommended that Dunn's suspension be lifted subject to restrictions, and the full Board again rejected the recommendation. In early 2000, after the full Board rejected a third ISC panel recommendation that Dunn should be reinstated, Dunn appealed the Board's decision. The matter was sent to the State Office of Administrative Hearings ("SOAH"). While the cause was pending with SOAH, Dunn and the Board participated in a mediated settlement conference. During that mediation, an agreement was reached under which Dunn's suspension would be lifted subject to various requirements and conditions. When the agreement was presented to the full Board, the Board rejected the agreement. The SOAH proceeding was then reopened. Following a lengthy hearing, the ALJ issued a proposal for decision, stating that the evidence established that Dunn was competent to again practice medicine and recommending that the Board reinstate Dunn's medical license subject to certain restrictions. The ALJ made extensive findings, most of which were not challenged or changed by the Board and which provide the underpinnings for the legal issues at hand. The Board rejected several of the ALJ's findings of fact and conclusions of law and refused to reinstate Dunn. Dunn appealed to the district court, which found that the Board's decision lacked evidentiary support; it reversed the Board's order and remanded it for further proceedings, restricting the Board to consideration of the record previously developed. The district court also ordered the Board not to bar Dunn from taking the Special Purpose Exam ("SPEX") and Jurisprudence Exam ("JE") to show his medical competence. It is from the district court's order that the Board and its Executive Director, Donald W. Patrick, M.D., J.D. (collectively, the "Board"), appeal. The Board raises two issues on appeal: (1) whether the Board had a reasonable basis to reject several of the ALJ's findings of fact and conclusions of law, and (2) whether the district court erred in limiting the scope of the Board's review of the cause on remand and in ordering the Board not to "prevent or hinder" Dunn's applications to take the SPEX and JE. Standard of Review The Board is the "primary means of licensing, regulating, and disciplining physicians," and is statutorily authorized to adopt rules to perform its duties and regulate the practice of medicine. Tex. Occ. Code Ann. §§ 151.003(2), 153.001 (West Supp. 2004). Once a license has been suspended, a doctor may seek reinstatement by showing that reinstatement is in the best interests of both the doctor and the public. Id. § 164.151(c) (West Supp. 2004). The Board's decision to deny reinstatement is subject to judicial review under the Administrative Procedure Act (the "APA"). Id. §§ 164.007(a), .009, .151(d) (West Supp. 2004); see Tex. Gov't Code Ann. §§ 2001.001-.902 (West 2000 & Supp. 2004). We review the Board's actions under the substantial evidence rule and will not substitute our judgment for that of the Board; we will reverse only if the decision is not reasonably supported by substantial evidence, is arbitrary or capricious, or is an abuse of discretion. Tex. Gov't Code Ann. § 2001.174(2)(E), (F) (West 2000). The Board must show it properly changed or disregarded the ALJ's findings and conclusions. The Board, like other agencies, does not have unlimited discretion to change an ALJ's findings of fact. Generally, an agency may not change an ALJ's findings of fact or conclusions of law or modify an ALJ's order unless the agency determines that (1) the ALJ improperly applied or interpreted the law, agency rules and policies, or prior administrative decisions, (2) a prior administrative decision on which the ALJ's decision was based was incorrect, or (3) a finding of fact contains a technical error requiring correction; the agency must state in writing specific reasons and legal bases for any changes. Id. § 2001.058(e) (West 2000); see Flores v. Employees Retirement Sys., 74 S.W.3d 532, 539-40 (Tex. App.--Austin 2002, pet. denied); see also Montgomery Indep. Sch. Dist. v. Davis, 34 S.W.3d 559, 564-65 (Tex. 2000) (once school board chooses to send dispute to hearing examiner, statutory scheme requires board to defer to fact findings and board may not reweigh evidence and resolve questions of credibility; statute is more restrictive than APA but agency retains authority to ultimately decide whether facts show violation of board policy); Southwestern Pub. Serv. Co. v. Public Util. Comm'n, 962 S.W.2d 207, 212-13 (Tex. App.--Austin 1998, pet. denied) (PUC has statutory authority to supplant ALJ's findings and is not bound by "the general, restrictive APA section 2001.058"). The Board, after receiving an ALJ's findings of facts and conclusions of law, is to "determine the charges on the merits." Tex. Occ. Code Ann. § 164.007(a) (West Supp. 2004). The Board argues that it is not bound by section 2001.058 because of this mandate to determine charges on the merits. But the fact that the Board must determine the merits of such cases is not inconsistent with the provisions of section 2001.058. This Court has held that section 2001.058 applies to this agency to delineate the allocation of fact-finding between the ALJ and the Board. Compare Levy v. Texas State Bd. of Med. Exam'rs, 966 S.W.2d 813, 815-16 (Tex. App.--Austin 1998, no pet.), with Southwestern Pub. Serv. Co., 962 S.W.2d at 212-13 (section 2003.049 expressly provides broader PUC review than section 2001.058); see also F. Scott McCown & Monica Leo, When Can an Agency Change the Findings or Conclusions of an ALJ?, 51 Baylor L. Rev. 63, 76-80 (1999) (discussing section 2001.058 and its application to agency decisions). Courts give deference to an agency's decision to modify or reject an ALJ's conclusions or findings only when substantiated by the ALJ's findings or when the basis for the change is fully articulated. Compare Ramirez v. Texas State Bd. of Med. Exam'rs, 995 S.W.2d 915, 918-19, 925 (Tex. App.--Austin 1999, pet. denied) (order followed ALJ recommendation; held substantial evidence supported board's decision), with Employees' Retirement Sys. v. McKillip, 956 S.W.2d 795, 800-01 (Tex. App.--Austin 1997), overruled in part on other grounds by Texas Natural Res. Conservation Comm'n v. Sierra Club, 70 S.W.3d 809, 814 (Tex. 2002) (after ALJ made findings of fact, agency made contrary findings and decision; held, agency did not satisfy section 2001.058 and did not show "rational connection between the stated policy and the change ordered by the agency"), and Texas State Bd. of Med. Exam'rs v. Scheffey, 949 S.W.2d 431, 432, 437 (Tex. App.--Austin 1997, writ denied) (board adopted ALJ's findings of fact but rejected recommendation; held, one of board's grounds for suspension was supported by substantial evidence). An ALJ, as an independent and impartial fact finder, is better suited to decide questions of so-called "adjudicative fact," meaning questions of fact affecting only the parties to a contested case, "the 'who, what, when, where and how' disputes of the case." On the other hand, agencies are "relatively" free to review and correct an ALJ's "legislative facts," which "provide a foundation for developing law, rules, or policies and, consequently, affect the outcome of many cases." McCown & Leo, supra, at 68-69 (citing K.C. Davis, Treatise on Administrative Law § 15.03, at 353 (2d ed. 1979)); see Exxon Corp. v. Railroad Comm'n, 993 S.W.2d 704, 710 (Tex. App.--Austin 1999, no pet.) ("'adjudicative' facts [are those] for which one expects perfectly explicit support in the body of evidence; . . . 'legislative facts' [are those] that the Commission was authorized to determine based on its judgment and expertise in light of what the evidence showed with respect to the general situation"); Kenneth Culp Davis & Richard J. Pierce, Jr., Administrative Law Treatise § 9.5, at 55 (3d ed. 1994) (distinguishing adjudicative facts from legislative facts); see also Davis, 34 S.W.3d at 565-66 (comparing more restrictive education code to APA; "the focus is on whether the issue determined is ultimately one of policy, and if so, whether a school board's decision is supported by substantial evidence and free of erroneous legal conclusions"). Thus, the allocation of primary fact-finding to the ALJ, with any changes to be appropriately substantiated by the Board, is a hallmark of the administrative process. Analysis To earn reinstatement of a suspended medical license, a petitioning doctor has the burden of showing that reinstatement is in his and the public's best interests. Tex. Occ. Code Ann. § 164.151(c) (West Supp. 2004). Once an ALJ finds that a doctor has made such a showing, the Board, if it disagrees, must explain how the ALJ's findings or conclusions were erroneous, based on the administrative record as presented to the ALJ. See Tex. Gov't Code Ann. § 2001.058(e). Here, the ALJ found that Dr. Dunn carried his burden, and the burden then shifted to the Board to controvert or establish why his reinstatement is not in his or the public's best interest. We hold that the Board did not carry this burden. The Board rejected the ALJ's findings that (1) before Dunn began abusing drugs he was clinically competent and (2) he was physically, mentally, and otherwise competent to safely practice medicine as long as he continued in recovery. (1) The Board refused to adopt the following conclusions of law: (1) Dunn established that it was in his and the public's best interest to have his license reinstated; (2) it would be appropriate to impose practice restrictions and monitoring to assure Dunn's continued recovery and clinical competence while he reentered the practice of medicine; and (3) Dunn should be reinstated with the requirements that he continue to submit to random urine tests, take a drug that blocks the brain's ability to feel a "high" from opiates, pass the SPEX and JE, complete a residency or fellowship, limit his practice to an institutional or group practice, and have his practice monitored by another doctor. The Board attempts to characterize the rejected findings as legislative facts, arguing that the determination of clinical competence impacts future cases. However, we believe the specific questions here--whether Dunn was competent before his drug abuse began and whether he is currently competent to practice medicine--are instead adjudicative facts, tied to and affecting only the parties to this particular contested case. The overall definition of competence is a matter of policy; whether a particular doctor under particular facts has carried his burden of establishing his competence is not. Therefore, we will review the Board's justifications for changing the ALJ's adjudicative facts under a stricter standard than we would review a determination or alteration of legislative facts. See Flores, 74 S.W.3d at 539-40; McCown & Leo, supra, at 68-69. Finding of Fact Number 38 The ALJ's finding of fact number 38 stated, "Before Dr. Dunn began abusing drugs, he was clinically competent." The Board refused to adopt that finding, instead stating that the finding contradicted findings in the 1997 Agreed Order and that Dunn's "failure to recognize or acknowledge that he did anything wrong in administering anesthetic to patient M.J. after more than four years of sobriety is a further indication of a lack of clinical knowledge or skill unrelated to his abuse of drugs." First, we note that on appeal and during the proceedings below, the parties agreed that Dunn's competence before his drug abuse began in 1995 was not at issue. Therefore, the Board had no justification to change the ALJ's finding related to his competence before his drug abuse began. Furthermore, the Board, which presented no evidence at the hearing, did not rebut the evidence demonstrating that Dunn was in fact competent before his drug abuse began. (2) The Board states that its change was supported by evidence of Dunn's responsibility for M.J.'s death, concerns raised as to his overall competence, his prescribing controlled substances to the woman with whom he was involved, and the physician's group's termination of its affiliation with him. However, M.J. died in July 1995, Dunn's physician's group terminated his affiliation and questions were raised about his competence during 1995, and Dunn prescribed controlled substances to the woman in late 1995 and early 1996. These issues arose during the time Dunn was abusing drugs and are not relevant to Dunn's competency before his drug abuse began. Dunn had practiced medicine since 1989 and was certified by the American Board of Medical Specialties in anesthesiology. Other doctors presented evidence that Dunn was competent until 1995, and Dunn testified that he had administered anesthetic in 4500 cases, including about 500 epidurals, and that only one, M.J.'s case, gave rise to a malpractice claim. We hold that the district court properly found that the Board lacked substantial evidence or a reasonable basis to change the ALJ's finding of fact number 38. Finding of Fact Number 57 Finding of fact number 57 found that based on the ALJ's other findings, Dunn is "physically, mentally, and otherwise competent to safely practice medicine, contingent on his remaining in recovery." The Board rejected this finding, stating that it was not supported by the evidence, based on an insufficient review of the evidence and unsound medical principles, and insufficient to protect the public. The Board quoted statistics related to relapses in doctors recovering from drug abuse and stated that Dunn had not proved his current competence, noting he had not practiced medicine for over four years and had not passed any objective, recognized test or completed an internship. The Board also stated that Dunn's testimony showed a "failure to recognize his error," indicated a lack of clinical ability, and contradicted other statements he had made in which he acknowledged mistakes in M.J.'s treatment, thus casting doubt on his credibility. Dunn objected to the admission of the Agreed Order into evidence, stating he did not believe that all the facts recited in the Order were true. Dunn testified that he signed the Agreed Order because, "I thought it was time for me to get on with my recovery and as--I signed it as a basis of, you know, coming to an agreement with the--with the Board where I could be on the process to getting my license back. I wasn't so concerned with the actual facts of it and there are several of them that--that are not true." Dunn said his father, who is also a doctor, advised him to sign the Agreed Order because it was the "only chance of getting my license back, and . . . that's why I signed it." The ALJ ruled that she would allow it as evidence of the allegations leveled against Dunn, not as Dunn's absolute admission of the truth of the allegations. The ALJ's proposed decision makes a similar recitation. (3) The Agreed Order recites that M.J. had an uncomplicated pregnancy, that Dunn, "in gross error, placed the epidural's catheter in her abdomen" and administered "inappropriate dosages" of anesthetic, and that M.J. and her unborn child died as a result of Dunn's actions and omissions. The Order further recites that Dunn failed to recognize M.J.'s deteriorating condition or to start lifesaving measures in a timely fashion. The autopsy report listed M.J.'s cause of death as complications of epidural anesthesia, specifically finding that (1) the catheter was within the retroperitoneum and (2) the circumstances suggested anesthetic toxicity. The level of anesthetic in M.J.'s blood was not increased, but time had passed from the anesthetic's administration and her time of death. The report stated that the temporal relationship of the administration of the anesthetic, the sequence of events preceding the cardiac arrest, and the absence of other findings that might explain the seizures and cardiac arrest, as well as the "epidural catheter's position within the abdomen," all "compellingly indicate" that M.J. died as a result of complications from the epidural. On cross-examination, the Board walked Dunn through the Agreed Order's findings related to M.J.; Dunn denied that M.J. had an uncomplicated pregnancy, saying instead that she had documented preeclampsia and was showing signs of it when Dunn started her epidural. Dunn said as he administered the epidural, he used a known technique to decide when the needle was properly placed and "every indication that I had that I've been trained to do to place an epidural indicated that it was in the correct spot." He testified that during the autopsy, "the catheter tip had been sheared off" and was found in M.J.'s abdomen. The medical examiner "asked me where the epidural catheter tip should be and so it's my opinion that he didn't even know where the epidural catheter was." Dunn said he had not seen objective evidence that he had misplaced the catheter, and he disagreed with several statements in the medical examiner's report. He further testified that he had given a test dose of anesthetic and had checked for symptoms that the epidural was incorrectly inserted; when no such symptoms appeared, he injected the anesthetic very slowly and carefully. He said patients with preeclampsia are prone to seizures that can lead to cardiovascular collapse, and stated that when M.J. went into seizures and cardiac arrest, he attempted to revive her to no avail. He said, "I did do everything that I possibly could and in a timely manner, and I am satisfied that I did everything that I possibly could and that I was very careful about the way I did the epidural on this lady. . . . I'm not trying to assess blame, but there were, you know, other circumstances that were not ideal." Dunn also said that the medical examiner's report found that M.J.'s blood level of anesthetic was within normal limits and not elevated, "which is important . . . because elevated levels of anesthetic can cause seizures." An unidentified consultant wrote a letter to the Board in 1996, (4) following a review of M.J.'s records, stating that the most likely cause of death was Dunn's failure to recognize that the catheter was improperly placed and a "relative" overdose of anesthetic. The consultant said M.J.'s medical records did not support Dunn's assertion that preeclampsia had escalated to eclampsia with seizures. The consultant said Dunn should not have administered more anesthetic after an initial dose gave no result, indicating an improperly inserted catheter. The consultant also faulted Dunn for not checking for resuscitation equipment before administering anesthetic. There was extensive testimony from several doctors involved in Dunn's recovery from drug abuse, testifying that Dunn had been in complete compliance with their recommendations and requirements, was in "full sustained recovery," and was at minimal risk for relapse. One of those doctors testified that, with regard to M.J.'s death, "I think he [Dunn] has accepted full responsibility for that. It's not possible to determine to what extent substance abuse played a role in that with, you know, a great deal of accuracy. . . . He does accept the understanding that we have of chemical dependency, that part of that process is recovering from its affects, being preoccupied with the drug to some extent, et cetera, and he accepts that as something that was potentially interfering with his normal functioning." Asked a second time about Dunn's drug abuse and its involvement in M.J.'s death, the doctor said he did not believe Dunn was using drugs or going through withdrawal symptoms at the time he administered anesthetics to M.J., but, "Dr. Dunn has accepted that the fact that he was abusing drugs at that time could play a role in any untoward event." Asked about his professional performance in relation to and at about the time of M.J.'s death, Dunn testified, "I feel like I--I did what I was trained to do and I--although I am--I have a deeper understanding of what substance abuse, how it can possibly have an affect on it. I do feel like that I handled the situation as I was trained to do and I handled it in as effective a manner as I could have." He said, "I'm comfortable with . . . how drug addiction can--can possibly affect one's clinical judgment, I--I realize that there--that it may have played a part in the unfortunate outcome of that malpractice case and I'm comfortable with that." On cross-examination, he stated, "I have come to realize that even though I wasn't acutely intoxicated or in active withdrawals, that substance abuse or alcohol abuse around that time can have an affect on someone's judgment. And I--while I'm willing to admit that, you know, there's a possibility that that--you know, that that's a possibility, I still feel like I--I--I handled the--a difficult situation to the best of my ability and training." He also stated that, although he believes he was practicing medicine to the best of his ability, "I've come to realize that, you know, substance abuse can affect one's judgment," and, "I agree that it's possible that I wasn't at my optimal best." Dunn continued to take medical education courses during his suspension, amassing 381 hours at the time of his ALJ hearing. Dunn testified that he also tried to stay informed about medical issues without practicing by reviewing medical cases for a law firm in Dallas. At the time of the hearing, Dunn had taken more than seventy drug tests, all of which were negative. He continues to take two or three random tests a month. Dunn testified that it was only the last agreed ISC recommendation that would have required him to take the SPEX and JE, for which he understood he had to have Board approval, and that he then began to request permission to take the exams. Dunn said he had not received Board permission to take those two examinations. The Board argues that Dunn did not show current clinical competence because he had not: (1) practiced medicine since his suspension in 1997, (2) taken or passed any recognized examinations, or (3) completed an internship or fellowship. However, had Dunn practiced medicine since his suspension, he would have been guilty of the unauthorized practice of medicine. See Tex. Occ. Code Ann. § 155.001 (West Supp. 2004). Dunn agreed to all conditions the various ISC recommendations would have imposed, including taking and passing the SPEX and JE, limiting his practice to a group or institutional setting, continuing to take random drug tests, being monitored by another doctor approved by the Board, limiting or abstaining from the practice of anesthesiology, and completing a six-month residency before resuming practice. The completion of a residency program and passage of the examinations would seem to answer the Board's concerns about Dunn's clinical competence and absence from the practice of medicine since 1997. The Board also points to section 163.11(a) of the administrative code, which requires that "[a]ll applicants for licensure shall provide sufficient documentation to the board that the applicant has, on a full-time basis, actively diagnosed or treated persons or has been on the active teaching faculty of an acceptable approved medical school within each of the last two years preceding receipt of an Application for licensure." 22 Tex. Admin. Code § 163.11(a) (2003). However, section 163.11(c) allows an applicant lacking such recent experience to obtain a license by meeting one or more conditions, such as certification by the American Board of Medical Specialties, passage of the SPEX, completion of continuing medical education hours, or a fellowship, mini-residency, or other structured remedial education. Id. § 163.11(c). The Board also used relapse statistics quoted by the ALJ to support its decision. The ALJ found, and the Board agreed, that a study indicated that 40 percent of anesthesiologists relapsed, versus 44 percent of other kinds of physicians, and that 81 percent of anesthesiologists had sustained recoveries longer than two years. The Board then states that a 40 to 44 percent chance that Dunn will relapse poses too great a risk to the public. Under this rationale, however, no anesthesiologist who abused opiates would ever be considered safe to be relicensed, because the statistics would always indicate at least a 40 percent possibility of relapse. The Board states that it doubts Dunn's competence because he has not practiced medicine recently, while refusing to reinstate his license, even with restrictions. The Board states that Dunn had not passed any examinations, a point which has now become moot, while failing to give him permission to take the SPEX or JE. The Board states that Dunn had not completed an internship or residency program, while refusing to approve the ISC recommendations that would require such a program's completion before Dunn could return to practice. Finally, the Board states that Dunn's refusal to accept responsibility for M.J.'s death indicates a lack of clinical competence and raises credibility issues because Dunn had previously accepted such responsibility. (5) Issues of credibility are to be decided by the impartial finder of fact, the ALJ, and the Board should not redetermine those issues or reweigh the evidence. See Flores, 74 S.W.3d at 539-40 (resolution of adjudicative facts "often requires making credibility determinations" and hearing examiner is better suited to make such determinations than agency because hearing examiner has heard evidence and observed witness demeanor and is disinterested party); Ford Motor Co. v. Motor Vehicle Bd., 21 S.W.3d 744, 757 (Tex. App.--Austin 2000, pet. denied) (quoting Southern Union Gas Co. v. Railroad Comm'n, 692 S.W.2d 137, 141-42 (Tex. App.--Austin 1985, writ ref'd n.r.e.)) ("The ALJ was the sole judge of the credibility of the witnesses and was free to accept the testimony of any witness or even accept 'part of the testimony of one witness and disregard the remainder.'"); Southwestern Pub. Serv. Co., 962 S.W.2d at 214 (Because PUC matters "often involve objective evidence that is more conducive to review on the record than evidence such as live witness testimony, which is subject to credibility concerns. . ., section 2003.049(g) allows the Commission to substitute its judgment for the ALJ's on questions of fact."). Further, Dunn, while refuting some of the allegations in the Agreed Order and stating that he believed other factors substantially contributed to M.J.'s death and that he had handled the situation as effectively as he could have, also testified that he understood that his drug abuse could have had an effect on his performance. We further observe that when the Board refused to reinstate Dunn's license after three ISC recommendations, Dunn had not yet testified before the ALJ, thus raising the asserted credibility concerns, and yet the Board denied each recommendation, even when they included requirements that would protect the public and assure Dunn's clinical competence. Based on the reasons put forth by the Board throughout these proceedings, it appears that the Board is not hewing to the statutory standard for reinstatement, but rather is acting consistent with the intent that Dunn shall never practice medicine again. (6) However, the Board entered into the Agreed Order, suspending Dunn's license until he could prove his competence, not revoking his license, as the Board is authorized to do. The Board should be bound by the Agreed Order to at least the same degree as it seeks to bind Dunn. The Board's reasons for changing the ALJ's findings of fact do not reference an improper application of law, policy, or rule, nor do they state that the ALJ relied on an incorrect prior decision or made a technical error. Thus, the Board was not authorized by statute to change the ALJ's findings of fact. See Tex. Gov't Code Ann. § 2001.058(e). Even if we assume that section 2001.058(e) does not apply, the Board's reasoning states that the ALJ's findings were not supported by the evidence, were based on insufficient review of the evidence and unsound medical principles, and were insufficient to protect the public. The Board opted not to present any evidence other than the Agreed Order, which Dunn disputed in some detail. The Board chose to rely solely on the Agreed Order and its cross-examination of Dunn's witnesses, and is therefore bound by the evidence as presented by those witnesses. That testimony indicates that Dunn has performed exceptionally well in his rehabilitation efforts, has maintained his recovery, and shows no signs of relapse. Dunn presented evidence of his clinical competence prior to his drug abuse, something the Board did not dispute other than to point to allegations of misconduct during the time Dunn was using drugs, and presented evidence of his efforts to maintain his competence by taking substantial numbers of continuing education courses and reviewing medical records for a law firm. Dunn agreed to complete a residency program, pass the SPEX and JE, and limit his practice. Dunn testified that he believed his administration of the anesthetic was not the sole cause of M.J.'s death, while admitting that his drug abuse may have affected his performance. The evidence does not support the Board's changing of the ALJ's finding of fact number 57. Conclusions of Law 5 through 9 In conclusions 5 through 9, the ALJ concluded that Dunn had shown it would be in his and the public's best interests for his license to be reinstated, that practice restrictions and monitoring requirements should be imposed to ensure Dunn's continued recovery and clinical competence, and that the Board should lift the suspension of Dunn's license subject to reasonable restrictions and requirements. As with the findings of fact discussed above, the Board justified its changes by stating the ALJ's conclusions were not supported by the evidence or based on insufficient review of the evidence and unsound medical principles, and were insufficient to protect the public, and by stating Dunn had "failed to meet his burden of proof that he is currently competent to practice medicine," again noting Dunn had not practiced in over four years, taken recognized tests, or completed a recognized internship or fellowship. We fail to see how a doctor whose license is suspended for a period of years could possibly demonstrate competence under the Board's interpretation of the requirements in this case. Based on the evidence as presented by Dunn and cross-examined by the Board and as discussed above, we hold that the reasons put forth by the Board do not justify the Board's changing of the ALJ's conclusions of law. Can the District Court Limit the Board's Review on Remand? We have held that on remand to an agency, a court may use its general equity powers to control the scope of remand or to provide instructions to the agency, as long as it acts within the bounds of the law. First Sav. & Loan Assoc. v. Lewis, 512 S.W.2d 62, 64 (Tex. Civ. App.--Austin 1974, writ ref'd n.r.e.). The APA authorizes a court to "reverse or remand the case for further proceedings." Tex. Gov't Code Ann. § 2001.174(2). We have held that under a similarly worded statute authorizing remand, a court may limit the scope of the remand to the record previously established by the agency. Texas Health Facilities Comm'n v. Nueces County Hosp. Dist., 581 S.W.2d 768, 770 (Tex. Civ. App.--Austin 1979, no writ); see Lewis, 512 S.W.2d at 64 ("the district court may control the scope of its remand"). The Board points to our ruling in BFI Waste Systems of North America, Inc. v. Martinez Environmental Group that a "reviewing court is empowered to issue only a general remand" under the statute in question. 93 S.W.3d 570, 579 n.9 (Tex. App.--Austin 2002, pet. denied). That case, however, is distinguishable because there the district court gave the agency "detailed instructions on how the commission should determine" the issue on remand. Id. Instructions to weigh the evidence in a certain manner conflict with the prohibition that a court shall "not substitute its judgment for the judgment of the state agency on the weight of the evidence." Tex. Gov't Code Ann. § 2001.174. We do not have such detailed instructions in this cause. (7) See also McKillip, 956 S.W.2d at 802 (holding trial court erred in remanding cause to agency with specific instructions, including instruction ordering agency to adopt proposal for decision). Limiting the scope of the record on remand does not direct the agency as to how it is to weigh the evidence and does not conflict with any statute. The Board had more than enough opportunity at the ALJ level to present evidence. The Board failed to provide sufficient justification for its decision to change the ALJ's findings and conclusions. See Tex. Gov't Code Ann. § 2001.058(e); Flores, 74 S.W.3d at 539-40. Given the circumstances and procedural history of this case, we hold that the district court did not err in limiting the Board's consideration to the ALJ record on remand. Conclusion After M.J.'s tragic death in 1995, the Board did not revoke Dunn's license, instead opting to suspend the license with conditions under which Dunn could seek reinstatement. We believe it is disingenuous for the Board to purport to allow Dunn to seek reinstatement and then essentially make it impossible for him to fulfill the Board's conditions by blocking him from taking required exams or by holding it against him that he has not practiced medicine while suspended or that he is resistant to take complete responsibility for M.J.'s death. Furthermore, the Board states that it denied Dunn's reinstatement in part because his testimony before the ALJ was inconsistent with prior testimony and the Agreed Order. However, the Board had already denied reinstatement three times before Dunn gave that "inconsistent" testimony, and more importantly, the ALJ is charged with resolving issues of credibility and evidentiary conflicts. We recognize the Board's ultimate responsibility of policing the medical profession to ensure the public's safety. Notwithstanding the importance of this goal, it is not inconsistent with the allocation of fact-finding between the ALJ and the Board and the requirement that the Board establish an evidentiary basis in the record for any modification of the ALJ's findings. We hold that the Board failed to carry its burden to articulate a reasonable evidentiary basis for rejecting the ALJ's findings of fact and conclusions of law. We affirm the district court's judgment reversing the Board's order and limiting the Board's review on remand to the record that was before the ALJ. __________________________________________ Jan P. Patterson, Justice Before Chief Justice Law, Justices B. A. Smith and Patterson Affirmed Filed: November 20, 2003 1. The Board also rejected several findings by the ALJ related to the SPEX and JE. However, at oral argument, both parties agreed that the issues related to Dunn's desire to take the examinations were moot. Therefore, we will not address the Board's complaints related to the ALJ's findings 44, 45, 46, and 47, and we will not address its complaints related to the district court's order barring it from interfering with Dunn's attempts to take the examinations. 2. At the hearing before the ALJ, counsel for the Board stipulated that "at some point prior to 95 Dr. Dunn was competent. As to the exact time period, we couldn't be specific as to what time period he was competent." 3. Because the Board and Dunn, who was not represented by an attorney and who signed the order without advice of counsel, opted to enter into the Agreed Order in 1997 and Dunn settled the malpractice suit, the actual facts of M.J.'s tragic death were never fully litigated. The Board wishes to use Dunn's signature to the Agreed Order as an absolute admission of all statements of fact in the order. However, the Board, by its decision to enter into the Agreed Order, chose not to fully explore the facts of the case. Dunn testified that he did not believe every factual recitation was true, but that he signed the document in order to get on with his recovery and his life. Having voluntarily entered into the Agreed Order, neither party may collaterally attack the Order. However, we recognize that the Order is in the nature of a compromise. We believe the ALJ properly admitted the Agreed Order, not as "admissions by Dr. Dunn or . . . equivalent[s] to adjudicated facts," but as an indication of the findings and events surrounding M.J.'s death and Dunn's drug abuse. It is probative but not utterly binding evidence. The ALJ took his testimony into account when evaluating his credibility and competence. The record reflects that the ALJ carefully considered Dunn's credibility, even reopening the SOAH hearings to address inconsistencies in testimony that raised credibility concerns. 4. This letter is unsigned and does not indicate by whom it was written, that person's credentials or qualifications, or his or her relationship to the Board or this case. It appears that it was never admitted during the ALJ's hearing, but instead was attached as an exhibit to one of the Board's pleadings. 5. The Board also refers to the allegations from 1995 and 1996. However, any concerns raised during the time Dunn was abusing drugs are irrelevant to his competence now. 6. This is further indicated by the minutes from the Board meeting at which the last ISC recommendation to reinstate Dunn's license was discussed. At that meeting, a Board member stated "it is his feeling that Dr. Dunn's actions were so egregious that he personally will not vote to approve his request to terminate his suspension." 7. As noted earlier, the parties agree that the issues related to Dunn's attempts to take the SPEX and JE are moot. We therefore do not consider those instructions in this discussion.