Opinion of the Court by
Justice ABRAMSON.
Judith Burton, a resident of Clay, Kentucky, brought suit against Dr. Philip Tro-ver and the Trover Clinic Foundation (TCF) in the Hopkins Circuit Court alleging that Dr. Trover, a radiologist, misread computed tomography scans (CT scans or CTs) of her lungs and thereby delayed the diagnosis of her lung cancer. She further alleged that TCF, Dr. Trover’s employer,1 was vicariously liable for Dr. Trover’s alleged negligence and was negligent itself in credentialing, i.e., granting staff privileges to, Dr. Trover. Prior to trial Judith Burton died, and her Estate revived the complaint only with respect to TCF. TCF thereupon impleaded Dr. Trover, and the matter went to trial in that posture. Following the three-week trial — four days of which were devoted to selecting the jury— and a jury verdict for Dr. Trover, the trial court dismissed all of the Estate’s claims. The Estate then appealed to the Court of Appeals. That Court, although affirming on all other grounds, concluded that the trial court erred by not allowing the Estate to cross-examine Dr. Trover regarding the status of his Kentucky medical license. Concluding further that the error was not harmless, the Court of Appeals reversed the trial court’s judgment and remanded for additional proceedings. Dr. Trover and TCF then sought discretionary review in this Court, which review we granted, in large part in order to consider the Court of Appeals’ recognition of a new cause of action against a hospital for “negligent credentialing.” As it happens, our review convinces us that the trial court’s evidentiary ruling was not an abuse of its [168]*168discretion, as held by the Court of Appeals, and so we are constrained simply to reinstate the trial court’s judgment and to leave for another day consideration of a negligent credentialing cause of action.
RELEVANT FACTS
Burton’s suit was one of some forty-nine brought against Dr. Trover in the wake of allegations by Dr. Neil Kluger, an oncologist at the Regional Medical Center’s Mahr Cancer Center, that Dr. Trover had poor practice habits and was not a reliable reader of mammograms and other diagnostic radiographic images. Dr. Kluger’s allegations, which commenced in early 2004, were addressed to the Medical Center’s Medical Executive Committee, to TCF’s Board of Governors, and to the Kentucky Board of Medical Licensure. The allegations were soon made public and received considerable attention from news media in the region — hence the trial court’s painstaking efforts to seat the jury.
The Medical Executive Committee investigated Dr. Kluger’s allegations by questioning physicians who made frequent use of the Medical Center’s radiology services, by questioning other employees in the radiology department, and by submitting selected imaging studies performed by Dr. Trover about which questions had been raised-both mammographic and non-mammographic studies — to an outside reader for review. In the Committee’s view, all of these lines of inquiry raised concerns that, although “well trained and capable,” Dr. Trover lacked “consistent diligence.” The Committee was also concerned by the fact that Dr. Trover typically interpreted more than 30,000 radiological examinations per year, whereas, according to one surveying group at least, the average workload for a full time radiologist is 12,800 per year. In April 2004, the Executive Committee recommended to the Board of Governors that Dr. Trover’s clinical privileges be revoked and his membership on the Medical Staff terminated, subject to reinstatement upon certain conditions.2 Subsequently, TCF undertook a randomized review of some 10,000 of Dr. Trover’s interpretations from 2003 and 2004, and that review, according to TCF, indicated that the rate of discrepancy between Dr. Trover’s interpretations and those of the reviewers was “well within the standard of care.”
The Kentucky Board of Medical Licen-sure, after receiving a grievance from Dr. Kluger, investigated the matter by contacting the Regional Medical Center to obtain additional information and by retaining consultants. This culminated in a Complaint against Dr. Trover and an Emergency Order of Suspension on July 14, 2005. By that point, Dr. Trover had not practiced in Kentucky for almost a year, having resigned from the Regional Medical Center in August, 2004 and resumed practice in Michigan. He contested the Board’s findings, maintaining that Dr. Kluger’s allegations were not true and were part of “a malicious effort to harm him” personally and professionally. He presented his own consultants, three Board-certified radiologists, who contradicted the Board’s consultants’ findings and concluded that Dr. Trover was competent and well-qualified. The matter was eventually resolved through an informal resolution process that produced an April 13, 2006 Agreed Order. That order was later replaced with a substantially similar Amended Agreed Order of May 9, 2007. The latter Order updated the status of the [169]*169matter by reflecting that Dr. Trover had completed an Education Plan outlined by the Board.
The May 2007 Amended Agreed Order is the licensure action that was in issue at trial. It addressed four major areas of inquiry: Dr. Trover’s reading of CT scans; his reading of mammograms; the volume of readings performed by Dr. Trover annually; and the intracranial interventional procedures he performed as a physician Board certified in radiology and interven-tional radiology. Only the reading of CT scans and the volume of readings are arguably relevant to this matter. The relevant portions of the Amended Agreed Order state as follows:
5. A Board Consultant, Board Certified in Radiology with Certifícate of Added Qualifications in Neuro-radiology, was retained to review diagnostic radiographic studies and interventional procedures performed by the licensee. The Consultant found the diagnosis in Patient A’s case, where the licensee missed an instance of recurrent colon cancer, to be below the expected standard of radiographic care and skills within the Commonwealth of Kentucky. The Consultant noted that in the case involving Patient B, the licensee reviewed the CT scan on more than one occasion and yet “failed to see a mass of 16 centimeters in diameter.” The Consultant added, “[although a precise identification of the type of mass is not possible, failure to appreciate a lesion of this size is far below the expected standard of radiographic care and skills within the Commonwealth of Kentucky.” In addition to patients A and B, the Board Consultant reviewed eight (8) additional diagnostic radiology cases. The Consultant only found a significant disagreement with the licensee’s reading of the images of Patient C. The Board Consultant found that the licensee missed a large pneumoperitoneum and hilar mass in Patient C....
The Consultant also found that the 33,000 radiological images read by the licensee in one year to be outside the norm for a busy practicing radiologist. The Consultant reported that in his own practice, the radiologist (sic) average 15,000-18,000 images read per year.
⅞: sjt ⅝ ⅝ ⅜ ⅜
7. The licensee provided both oral and written responses to the grievances during the course of the investigation. In response to allegations pertaining to Patient A, the licensee stated that had he been aware of the patients history of previous neoplasm, it would have assisted him in the interpretation of the exam and he would have looked primarily for metastatic disease. The licensee acknowledged the missed diagnosis but maintained that “this was an honest miss, not due to lack of training, effort or other fault except a perceptual mistake,” and “when he saw the study later, he immediately recognized the metastatic deposit.” The licensee emphasized that the patient was not harmed and her prognosis was not affected by this mistake.
In Patient B’s case, the licensee denies missing the abscess, “because it did not exist.” The licensee contends that a CT-scan three days before surgery showed no abscess, nor was (sic) there any clinical signs of harboring an abscess exhibited by the patient. The abscess appears to have been, “the result [170]*170of surgery,” or “probably formed after surgery due to a bile leak.” The licensee asserts that this is not uncommon in this type of procedure.
The licensee is Board certified in both Radiology and interventional radiology. The licensee reported reading approximately 33,000 radiological cases during 2003 due to insufficient radiology staffing at RMC. The licensee maintains that he practiced within the American College of Radiology workload averages during his other 24 years of practice.
⅝ # ⅜ ⅝ ⅝ ⅝
8. The Board’s first consultant provided testimony in which he conceded that 3 diagnostic cases (Patients A, B and C) do not constitute a fair sampling to determine the competence level of a radiologist, especially when the cases (Pat. A and B) are not taken at random.
* # * * * *
9. The licensee retained as witnesses, two (2) Board Certified Radiologists who disagree with the Board’s 1st consultant regarding Patient “A,” and find that no abscess was in existence when Patient A’s CT was read by the licensee, and that the image described by the Board’s 1st Consultant is actually Patient A’s uterus and displaced bladder. Another witness for the licensee is the medical examiner who performed the autopsy on Patient A and finds that the abscess noted in the autopsy, based on his anatomical review, was not present twelve days before her death when the licensee read the CT and therefore he did not miss the mass.
⅜ ⅜: ⅜ ⅜ ⅜ i|:
11. The licensee retained as witnesses, three (3) Board Certified radiologists, including the Chairman of the Department of Radiology at Kosair Children’s Hospital, who all find that 3 diagnostic cases (Patients A, B and C) are an inadequate sampling to determine the licensees competence; that in reviewing the Board’s investigation as a whole, the licensee has committed no acts of gross negligence, gross malpractice, or unprofessional conduct, nor has he done anything to bring the medical profession into disrepute; that the licensee is a well-qualified, competent radiologist who demonstrates good clinical judgment and exercises appropriate patient management decisions.
12. The Trover Foundation’s Regional Medical Center, by letters of August 10, 2005 and August 22, 2005 to the Board reported that the allegations raised by Dr. Kluger were investigated by their hospital and found to be largely unsubstantiated. Further, the hospital re-read more than 10,000 of the licensee’s films and determined that he was well within the standard of care.
The Board thus outlined its consultant’s findings, acknowledged the countervailing expert testimony and other evidence proffered by Dr. Trover, the small scale of its own review, Dr. Trover’s compliance with a minor educational recommendation by the Center for Personalized Education for Physicians, and his denial of any wrong doing. In light of the agreement between the parties, the Board made no findings beyond the parties’ stipulation that cause existed for entry of the April 2006 (and ultimately May 2007 Amended) Agreed Order, the gist of which is that Dr. Tro-ver’s license to practice medicine in Kentucky would for an indefinite period be conditioned on his not interpreting more [171]*171than one hundred images per actual day worked and on the regular review of his interpretations by a radiologist approved by the Board. Both Agreed Orders provide that the pending Complaint was finally resolved through “an informal resolution” pursuant to KRS 311.591(6) and 201 KAR 9:082.
In the midst of this controversy, Burton’s complaint, after numerous amendments, focused finally on three instances when, she alleged, Dr. Trover studied CT scans of her lungs and failed to identify and report lesions, one in either lung, that ultimately, she claimed, became cancerous. Her experts testified to the effect that the lesions were identifiable as matters of concern as early as February 2003, when Dr. Trover made the first of his alleged misreadings, and that treatment at that point would have been successful. As it happened, however, Burton’s doctors did not diagnose her with lung cancer until August 2004 — several months after Dr. Trover’s alleged misreadings in September 2003 and January 2004 — when the biopsy of a mass in Burton’s right lung discovered cancer cells. That mass and a mass in her left lung, the biopsy of which was inconclusive, were treated with radiation. Following that treatment, Burton was, for a time, deemed cancer free, but in late 2006 a new lung carcinoma was discovered. She died in 2008. According to her Estate’s expert, she would not have died at that time had her cancer been earlier detected and treated.
Dr. Trover introduced expert testimony (the head of the radiology department at the Vanderbilt University Medical Center and a pulmonary oncologist from the University of Kentucky Medical Center) to the contrary. These experts testified that pri- or to August 2004, Burton’s CT scans did not indicate the presence of lung cancer; that Dr. Trover had not misread the scans of February 2003, September 2003, and January 2004; that other radiologists at the time had concurred in Dr. Trover’s readings; and that Burton, who .had smoked from a young age and who suffered from severe chronic obstructive pulmonary disease (COPD), would not have fared better had she been treated earlier. The jury, as noted, found that Dr. Trover had not breached his duty of care, a finding that obviated any further proceedings with respect to Burton’s claims against TCF.
During the trial, Dr. Trover testified on his own behalf, and prior to his taking the witness stand he moved in limine to exclude any questions concerning,, and any reference to, the fact that his Kentucky medical license had been temporarily suspended and later subjected to the conditions reflected in the Agreed Orders. He argued that his licensure status was a collateral matter irrelevant to his performance in this particular case. Burton3 responded by noting that Dr. Trover would be testifying in two capacities — as an expert and as the defendant. As an expert, she asserted, Dr. Trover was subject to impeachment with the fact that his Kentucky license had been restricted. This evidence was also directly relevant to her claim and to Dr. Trover’s defense, she maintained, because it was based in part on the Board of Medical Licensure’s expressed concern that Dr. Trover had been interpreting an excessive number of radio-logic films per year. Burton had been permitted during her case in chief to introduce evidence tending to show that Dr. Trover had a “habit” of reading films rap[172]*172idly, that for several years, including 2003 and 2004, he had read more than 30,000 films per year, and that 30,000 films per year was well in excess of the “average” radiologist’s workload. The Board’s having conditioned Dr. Trover’s license on his not reading more than one hundred films per actual day worked was relevant, Burton claimed, to this “habit” line of testimony. The trial court acknowledged that the licensure matter and the reasons for it might have some relevance, but it ruled that the probative value of that evidence would be far outweighed by its prejudicial effect, and so sustained Dr. Trover’s motion to exclude it.
Reversing on this issue, the Court of Appeals took a slightly different tack. In its view the license status evidence was relevant and was a fair subject for cross examination because the Board proceeding had been prompted in part by allegations of a misread CT scan (although not any of Burton’s CTs) and because the Board’s Order conditioned Dr. Trover’s license, in part, on the regular review of a sampling of his CT-scan readings. The Board’s concern with allegations that Dr. Trover had misread other CT scans, the Court of Appeals believed, was both temporally proximate and directly relevant to Burton’s allegations that he had misread her CT scans, and so should not have been excluded as “collateral.”
Dr. Trover and TCF argue that the Court of Appeals’ analysis is incomplete and fails to consider the principal reason the trial court gave for excluding the license status evidence. As they correctly note, the trial court did not base its ruling simply on a determination that the evidence was collateral. It found, rather, that the evidence, although relevant to some extent, would prove far more prejudicial than probative. The Court of Appeals’ failure to take that balancing into account, Dr. Trover and TCF maintain, renders its decision unsound and has the effect of usurping the trial court’s discretion. We agree.
ANALYSIS
Kentucky Rule of Evidence 404(b) provides that “[e]vidence of other crimes, wrongs, or acts is not admissible to prove the character of a person in order to show action in conformity therewith.” The rule applies to civil as well as to criminal proceedings, Ky. Farm Bureau Mut. Ins. Co. v. Rodgers, 179 S.W.3d 815 (Ky.2005), and its purpose is to guard against the substantive use of so-called character or propensity evidence. This type of evidence is generally evidence that on other occasions a person has acted in a particular way, and it is offered as proof that the person, being the sort of person who does that sort of thing or acts that way, is likely to have done the same sort of thing or acted that same way on the occasion at issue in the case. Our courts have long been concerned that triers of fact are apt to give such evidence more weight than it deserves, and that such evidence poses a substantial risk of distracting the trier of fact from the main question of what actually happened on a particular occasion. Clark v. Commonwealth, 223 S.W.3d 90, 96 (Ky.2007) (“Ultimate fairness mandates that an accused be tried only for the particular crime for which he is charged.”) (citation and internal quotation marks omitted); Robert G. Lawson, The Kentucky Evidence Law Handbook, § 2.40[3], at 183, 184 (4th ed.2003) (noting the long line of pre-Rules cases to the effect that “other negligent acts of a person may not be used to prove that he or she was negligent on a given occasion”). Dr. Trover’s alleged misreadings of other patients’ CT scans and the resulting restriction of his license come within this rule and, as the trial court recognized, were not admissible [173]*173as substantive evidence to show that Dr. Trover misread Burton’s scans. See Bair v. Callahan, 664 F.Sd 1225 (8th Cir.2012) (holding that surgeon’s past treatment of other patients was not admissible to show that he negligently treated the plaintiff); Ky. Farm Bureau Mut. Ins. Co. v. Rodgers, 179 S.W.3d at 815 (holding that insurance company’s prior allegedly unfair claim settlement was not admissible to show that it unfairly settled the plaintiffs claim).
Although KRE 404(b) precludes the use of evidence of collateral misconduct to show a person’s propensity to engage in such conduct, the rule also provides that evidence of other wrongs or acts may be admissible “if offered for some other purpose, such as proof of motive, opportunity, intent, preparation, plan, knowledge, identity, or absence of mistake or accident.” KRE 404(b)(1). Even if relevant for some other purpose, however, Rule 404(b) evidence is also subject to KRE 403, which authorizes the trial court to exclude relevant evidence “if its probative value is substantially outweighed” by the risk of some harm, such as undue prejudice or confusion of the issues. Lanham v. Commonwealth, 171 S.W.3d 14 (Ky.2005). In Bell v. Commonwealth, 875 S.W.2d 882 (Ky.1994), this Court in effect amalgamated the two rules and held that Rule 404(b) evidence is admissible only if (1) it is relevant for a legitimate purpose; (2) it is probative, i. e., only if there is sufficient evidence that the other crime, wrong, or act actually occurred; and (3) its probative value is not substantially outweighed by any prejudicial effect. Id. at 889-91. We review a trial court’s application of these requirements for abuse of discretion. Purcell v. Commonwealth, 149 S.W.3d 382, 400 (Ky.2004); Commonwealth v. Prater, 324 S.W.3d 393, 398 (Ky.2010). The question then is whether evidence of Dr. Trover’s license restriction and the allegations before the Board of Medical Licensure that he misread other patients’ CT scans so clearly satisfied these requirements that it was an abuse of the trial court’s discretion to rule otherwise.4
With respect to the first requirement, relevancy, Burton originally offered two purposes other than proof of propensity for which the license-status evidence could be deemed relevant. It served, she [174]*174claimed, to impeach Dr. Trover’s credibility as an expert witness, and substantively it added somewhat to the proof she had introduced during her case in chief to the effect that in 2003 and 2004 Dr. Trover read far more radiographic films than did an “average” radiologist and that he habitually read films rapidly. The Court of Appeals based its ruling on its view that Burton should have been allowed to impeach Dr. Trover’s testimony, his credibility, with both the fact that his license had been restricted and the reasons therefore.5
We agree with Burton and with the Court of Appeals that impeachment is a purpose other than propensity for which collateral act evidence might be relevant. Under KRE 608, indeed, even character evidence may be admissible in certain forms for impeachment purposes, but that rule limits the attack to evidence of a witness’s character for truthfulness or veracity, and there is no claim here that Dr. Trover’s license status is evidence of that sort. Otherwise, litigants are entitled to introduce extrinsic evidence to contradict a witness’s testimony on matters that are or have been made material to the merits of the case. Lawson, supra at § 4.05[1] p. 270 (“Needless to say, contradiction on material facts is a perfectly proper and acceptable form of impeachment.”). However, “[t]he general rule is that a witness cannot be cross-examined on a collateral matter which is irrelevant to the issue at hand.” Morrow v. Stivers, 886 S.W.2d 424, 429 (Ky.App.1992); Purcell v. Commonwealth, 149 S.W.3d at 397-98 (“Although there is no provision in the Kentucky Rules of Evidence prohibiting impeachment on collateral facts, we have continued to recognize that prohibition as a valid principle of evidence.”).
Unfortunately, it is not always easy to say whether, for the sake of contradiction, a matter should be deemed material or collateral. Lawson, supra § 4.05[2] p. 272 (noting that application of the rules governing impeachment by contradiction is made difficult “because of the complexity involved in determining ‘collateralness’”). On the one hand are the risks of issue proliferation and distracting the jury from the main issues in the case, Lawson, supra, § 4.05[4] p. 276, but on the other hand, courts are loath to allow a witness on direct examination to engage in perjury, mislead the trier of fact, and then shield him or herself from impeachment by asserting the collateral fact doctrine. Prater, 324 S.W.3d at 397 (holding that the trial court did not. abuse its discretion by admitting extrinsic evidence contradictory of matter defendant raised on direct examination meant to cast herself in a sympathetic light). As the Court observed in United States v. Castillo, 181 F.3d 1129, 1132-33 (9th Cir.1999), direct-examination testimony containing broad disclaimers of misconduct can sometimes “open the door for extrinsic evidence to contradict even though the contradictory evidence is otherwise inadmissible under Rules 404 and 608(b).”
What Burton claims, and what the Court of Appeals in effect held, is that Dr. Tro-ver’s testimony with respect to his qualification as an expert — he testified at some [175]*175length regarding his education, his board certification, and his experience as a radiologist at the Trover Clinic — was meant to cast himself in a favorable light and to give weight to his opinions — he testified that he did not misread Burton’s CT scans — and thus opened the door to “contradiction” by way of the license-status evidence. To so hold we would have to understand “contradiction” in a very loose sense, because there is no suggestion here that Dr. Trover misrepresented his credentials in any way, and he did not claim to be licensed in Kentucky, in good standing with Kentucky’s licensing authorities, or never to have misread a CT scan. The license-status evidence, in other words, does not contradict Dr. Trover’s testimony in any literal sense.
In Morrow v. Stivers, 836 S.W.2d at 424, the Court of Appeals, and in Reece v. Nationwide Mut. Ins. Co., 217 S.W.3d 226 (Ky.2007), this Court addressed similar claims that an opposing party should have been allowed to impeach a medical expert with evidence regarding his license. In both cases, unlike this one, the licensure board had actually taken action and suspended the physician’s license following an investigation. Although in both cases the Court held that the evidence ran afoul of the prohibition against impeachment on collateral facts, we explained that holding in Reece by noting that “in both cases the reason for license suspension had no relation [temporal or factual] to the case in which they [the experts] were testifying and was likely to be highly inflammatory.” 217 S.W.3d at 232. This explanation could perhaps be read to suggest that a license suspension that was related temporally and factually to the case in which the expert was testifying would be admissible. That is obviously how the Court of Appeals understood it. Ignoring the fact that Dr. Trover’s license was not suspended following the Board’s investigation but rather restricted by the Amended Agreed Order, the Court of Appeals concluded that Dr. Trover’s license “suspension” and the reasons underlying it should have been fair game during Dr. Trover’s cross-examination. The Court reasoned that the Board’s allegations that he misread CT scans at about the time that he read Burton’s scans was temporally and factually related to his testimony in Burton’s case
The Court of Appeals’ ruling is akin to that of the Supreme Court of South Dakota in Kostel v. Schwartz, 756 N.W.2d 363 (S.D.2008). The defendant doctor in that case, a neurosurgeon, was accused of misreading x-rays and conducting surgery beyond the patient’s consent. The South Dakota Supreme Court held that the trial court had not abused its discretion by ruling in limine that if the defendant qualified himself as an expert by testifying as to his education, training, and experience and testified as an expert regarding the standard of care, he would open the door to cross-examination concerning the suspension of his license as a result of two other cases in which, by his own admission, he had misread x-rays and performed back surgery at the wrong level of the patient’s spine. Nor had the trial court abused its discretion under Rule 404(b), the South Dakota Supreme Court held, by allowing the defendant to be asked during cross-examination three questions regarding the other instances of misread x-rays and inappropriate surgeries. The prior act evidence was relevant to the defendants knowledge, in the Courts view, and not merely to his propensity to commit negligence.
With respect to this latter point, we share the Eighth Circuit Court of Appeals’ concern that in many if not most medical negligence cases the defendant doctor’s knowledge is not genuinely at issue. Bair, 664 F.3d at 1225 (holding that defendant [176]*176surgeon’s mistreatment of other patients was not admissible as evidence of “knowledge” under rule 404(b)). In this case, for example, Dr. Trover’s education, training, and experience gave him the requisite knowledge. Indeed, the Amended Agreed Order states that the Center for Personalized Education for Physicians (CPEP) “found that the licensee’s knowledge was complete in the majority of diagnostic and interventional topics discussed.” What was at issue, in this case, was Dr. Trover’s competency, and evidence that he misread the CT scans of other patients thus ran not to his “knowledge” but merely to his propensity to commit malpractice. To the extent, therefore, that the Court of Appeals’ ruling can be understood as suggesting that there was a legitimate substantive purpose for admitting evidence of Dr. Tro-ver’s license status, we disagree.
Whether a factually and temporally related license suspension might be relevant to impeach an expert’s credibility is a harder question. In a number of cases we have upheld the admissibility of evidence with a substantial tendency to impeach expert testimony. See, e.g., Kemper v. Gordon, 272 S.W.3d 146 (Ky.2008) (holding that opposing party should have been allowed to question expert about inconsistent testimony he had given in another case); Ky. Farm Bureau Ins. Co. v. Rodgers, 179 S.W.3d at 821 (holding that evidence contradicting the asserted basis for the expert’s opinion was admissible as impeachment notwithstanding the fact that the evidence also related to the defendant’s prior act); Tuttle v. Perry, 82 S.W.3d 920 (Ky.2002) (holding that opposing party should have been allowed to ask expert how much he was being paid to testify). In light of these cases we are unwilling to say that an expert’s license suspension will always be collateral and irrelevant.
Even if an expert’s license suspension, or as in this case, restricted license is closely enough related to the expert’s testimony in the given case to be relevant, nevertheless, as the trial court concluded here, its admissibility still depends on the other two Bell requirements, ie., on whether there is sufficient evidence that the alleged prior acts of misconduct actually occurred and on whether the probative value of the prior-act evidence would be substantially outweighed by its prejudicial effect. The portion of our Reece opinion quoted above makes this point, implicitly at least, by noting that an important reason for excluding the license-suspension evidence in that case and in Stivers was the fact that the grounds for the suspensions in those cases were apt to be highly inflammatory.
Here, the Court of Appeals appears to have disregarded these other Bell requirements, even though the trial court based its ruling on its view that the prejudicial effect of permitting evidence concerning Dr. Trover’s license status would far outweigh the probative value of that evidence. The trial courts ruling, we are convinced, was well within its discretion. In the first place, while there is no dispute that Dr. Trover’s license was in fact restricted, it was not suspended following proceedings before the Board. As noted above, Dr. Trover never admitted the alleged wrongs giving rise to the licensure matter. On the contrary, before the Licensure Board he presented expert testimony to the effect that he had not misread the" CT scans, as was alleged. The Amended Agreed Order imposing restrictions, moreover, did not purport to find as a fact that Dr. Trover had misread CTs or made any other errors but was instead specifically described as “an informal resolution.” Indeed, on balance the Amended Agreed Order contains as many, if not more, findings that are either neutral or favorable to Dr. Trover [177]*177than it does findings adverse to him. In the end, the Amended Agreed Order provided merely that certain restrictions were justified in light of the parties’ stipulation that there was evidence raising significant concerns about Dr. Trover’s practices.
Given Dr. Trover’s denials and the qualified nature of the Amended Agreed Order, the trial court would have had very reasonable concerns that introduction of the li-censure evidence, and specifically the Amended Agreed Order, could and likely would lead to a complicated trial-within-the-trial as to whether the alleged collateral CT-scan misreadings ever happened. Cf. Kostel, 756 N.W.2d at 373 (distinguishing the case before the court, in which the license suspension was based on admitted prior acts of faulty treatment, from another case—Boomsma v. Dakota, Minnesota Eastern R.R. Corp., 651 N.W.2d 238in which evidence of a license suspension based on mere allegations had properly been excluded); see also Bair, 664 F.3d at 1230 (trial court properly excluded evidence regarding defendant-surgeon’s treatment of other patients where it would unfairly prejudice the defendant “confuse the jury, and cause undue delay.”)
The trial court was also clearly cognizant that distinguishing this case from cases such as Stivers and Reece, was the fact that Dr. Trover was testifying not merely as an expert witness but primarily as the defendant, making the KRE 404(b) concerns especially significant. Burton is correct that Dr. Trover testified as to his expert qualifications, offered expert testimony interpreting Burton’s CT scans, and maintained that he had not misread them. His testimony, however, appears to have been in compliance (Burton made no objection to the contrary) with the trial court’s pretrial order limiting the testimony of treating physicians to issues relating to their personal knowledge of Burton’s diagnosis and prognosis and to their personal involvement in her care and treatment. This pretrial ruling was requested by Burton, entered by the trial court and, based on our review, not violated. Dr. Trover did not, for example, appeal to studies or to any other outside sources to support his interpretations of Burton’s scans, nor did he purport to say what was or was not within the standard of care.6
Having limited the scope of the “expert” portion of Dr. Trover’s testimony, the trial court could reasonably believe that relatively little expert testimony remained to be impeached. On the other hand, the risk of substantial 404(b) prejudice against Dr. Trover, as the defendant, from the license proceeding and underlying allegations of similar CT scan misreadings was great — precisely the sort of unfair prejudice the court had taken four days of jury selection to avoid. In Purcell v. Commonwealth, 149 S.W.3d at 382, we reversed a conviction for promoting the sexual performance of a minor because the trial court had erroneously admitted evidence of the defendant’s prior acts of “homosexual voyeurism.” Rejecting the Commonwealth’s argument that the prior-act evidence was admissible to impeach certain denials the defendant had made during cross-examination, we noted Professor Lawson’s suggestion that issues concern[178]*178ing impeachment by way of collateral facts be decided by applying the KRE 403 balancing test. That balance, we held, clearly tipped against the Commonwealth, and we observed that, at least with respect to a defendant, “[i]t would be a rare occurrence ... when the prejudicial effect of evidence of ‘other bad acts’ would not substantially outweigh the impeachment value of such evidence.” 149 S.W.3d at 398. The same can be said here and, indeed, with more force because the Amended Agreed Order has limited probative value,’ there being no definitive findings of prior bad acts. Clearly, under these circumstances, the trial court did not abuse its discretion by declining to find that this was that “rare occurrence” where the impeachment value substantially outweighed the prejudicial effect.
CONCLUSION
In sum, although we do not rule out the possibility that a license suspension could provide a valid means of impeaching an expert witness, there was no sound basis for admitting the license-status evidence at issue in this case. Following the completion of the Board of Medical Licensure’s investigation, there was no actual license suspension, simply an informal resolution of the proceeding by an Agreed Order that restricted Dr. Trover’s license. The trial court did not abuse its discretion by excluding evidence with such limited impeachment value against Dr. Trover, given the potential for confusing the issues to be tried and the strong likelihood that it would cause unfair prejudice. We reverse, accordingly, the Opinion of the Court of Appeals, and hereby reinstate the Judgment of the Hopkins Circuit Court.
All sitting. All concur.