United States v. Shinderman

*9 SELYA, Senior Circuit Judge. The admonition “physician, heal thyself’ is a biblical proverb, Luke 4:23 (King James), suggesting that people should address their own failings. In the case at hand, this admonition has both literal and figurative application. The tale follows. Mark S. Shinderman, M.D., is a physician specializing in psychiatry and the treatment of addiction. The federal government indicted him on a gallimaufry of criminal charges stemming from his unauthorized use of another doctor’s name and Drug Enforcement Administration (DEA) registration number. A jury convicted him on many of those charges. On appeal, the defendant challenges both his convictions and his sentence. He advances four assignments of error, which impugn (i) the district court’s refusal to apply the exclusionary rule to evidence allegedly obtained in violation of federal regulations designed to protect the confidentiality of substance abuse treatment records, (ii) the foreclosure of a proposed entrapment defense, (iii) an evidentiary ruling, and (iv) the use of an obstruction of justice enhancement during sentencing. Concluding, as we do, that each and all of these animadversions lack merit, we affirm the judgment below. I. BACKGROUND We rehearse here only those facts that are necessary to place this appeal in perspective. Other facts are added in our subsequent discussion of particular issues. Following a dramatic increase in drug-overdose deaths in Maine, the United States Department of Health and Human Services (HHS) began investigating possible links between this spike in mortality and health-care clinics specializing in the treatment of drug addiction. One of the providers scrutinized in the course of this administrative inquiry was CAP Quality Care, a high-dose methadone treatment clinic located in Westbrook, Maine. The defendant, an advocate of the combined use of methadone and benzodiaze-pine in the treatment of patients addicted to opiates, served as CAP’s sponsor and national medical director. Federal law stipulates that any doctor who prescribes a controlled substance must possess a state-specific DEA registration number. See 21 U.S.C. § 822(a), (e). When CAP opened in 2001, the defendant had only an Illinois medical license and DEA number (although he had obtained a temporary license to practice medicine in Maine). In the course of HHS’s administrative inquiry into CAP’s affairs, its investigators took a hard look at the defendant’s prescription-writing practices. The defendant apparently attempted to write prescriptions for controlled substances using his Illinois DEA number. After Maine pharmacies refused to fill those prescriptions, he started writing benzodiazepine prescriptions under the name and Maine-specific DEA number of Dr. Steven Keefe, a fully credentialled physician affiliated with CAP. At that point, the inquiry intensified. The government’s next step was to subpoena CAP’s Medicaid records. Under statutes and regulations governing the disclosure of substance abuse treatment records, the government was required to obtain court authorization to use the subpoenaed documents for prosecutorial purposes. See 42 U.S.C. § 290dd-2(b)(2)(C); 42 C.F.R. § 2.66(a). On April 24, 2003 — shortly after the government launched a criminal investigation aimed at the defendant — a federal magistrate judge granted an ex parte motion seeking leave to disclose to federal law enforcement personnel the Medicaid records garnered from CAP pursuant to *10 HHS’s original administrative subpoena. On August 22, the magistrate judge granted a second ex parte motion seeking authorization to release CAP patient treatment records to the prosecutors. On September 5, the magistrate judge granted yet a third ex parte motion; this motion sought access to medical records seized in the course of executing a search warrant. In each instance, the magistrate judge found good cause for disclosure and honored the government’s supplication that he delay notice of the order granting the motion to avoid compromising the criminal investigation. To that end, the magistrate judge maintained the orders under seal and decreed that notice to the defendant and all other affected parties could be deferred for up to ninety days. 1 The record contains copies of the notices subsequently sent by the government to the defendant as well as an affidavit from an HHS official attesting that notice of the issuance of the disclosure orders had been given within the allotted time frame. The dénouement came much later. On August 25, 2005 — the same date on which the government filed a civil action against CAP for, among other things, Medicaid fraud — a federal grand jury handed up a sixty-eight count indictment. The first twenty-five counts charged the defendant with unlawful use of a DEA registration number issued to another physician, in violation of 21 U.S.C. § 843(a)(2). Counts 26 through 50 charged him with aiding and abetting the acquisition of a controlled substance by misrepresentation, in violation of 21 U.S.C. § 843(a)(3). Counts 51 and 52 charged him with falsifying pharmacy records, in violation of 21 U.S.C. § 843(a)(4)(A). The last sixteen counts charged him with making false statements related to the delivery of or payment for health-care benefits, in violation of 18 U.S.C. § 1035(a)(2). The defendant filed a host of motions, seeking among other things to suppress the fruits of the administrative subpoenas and the search. Pertinently, he argued that all of the evidence disclosed in pursuance of the ex parte court orders should be excluded because the government had flouted the notice requirements imposed by federal substance abuse treatment regulations. The district court found that the reprieve for the giving of notice granted by the magistrate .judge had no legal basis and that the defendant should have been given notice as soon as the patient records were disclosed to prosecutors. United States v. Shinderman, 432 F.Supp.2d 149, 154 (D.Me.2006). The court nevertheless refused to suppress the evidence, reasoning that suppression was too drastic a remedy. Id. The trial lasted for nearly two weeks. A parade of CAP patients testified to having seen the defendant write Keefe’s name and DEA number on controlled substance prescriptions. The .centerpiece of the government’s case was Keefe’s testimony that, while he agreed to provide the defendant with pre-signed prescription blanks, he never authorized the defendant to sign his name or use his DEA number. In the defense case, the defendant testified that Keefe had given him such permission. After a day and a half of deliberations, the jury found the defendant guilty on the twenty-five counts that alleged unlawful use of a DEA number belonging to another physician, twenty-four of the twenty-five aiding and abetting counts, both counts dealing with falsification of pharma *11 cy records, and seven of the sixteen false statement counts. The government voluntarily dismissed the remaining aiding and abetting count, and the jury acquitted the defendant on nine of the false statement counts. The sentencing phase followed. The district court applied enhancements for obstruction of justice and abuse of a position of trust, and set the guideline sentencing range (GSR) at 15 to 21 months. The court then departed downward and imposed a six-month incarcerative sentence. This timely appeal followed. II. ANALYSIS We separate the defendant’s assevera-tional array into its four component parts and treat those parts sequentially. A. Suppression. In the interest of guaranteeing the efficacy of substance abuse treatment programs, Congress has directed that all substance abuse treatment records be kept confidential. See 42 U.S.C. § 290dd-2(a). Disclosure is permitted only in narrow sets of circumstances. These include consent, id. § 290dd-2(b)(l); audit, id. § 290dd-2(b)(2)(B); and court order, id. § 290dd-2(b)(2)(C). Applicable regulations confirm and elaborate upon the statute’s commitment to confidentiality. See 42 C.F.R. §§ 2.1-2.67. The regulations warn that such records “may be disclosed or used only as permitted by these regulations and may not otherwise be disclosed or used in any civil, criminal, administrative, or legislative proceedings conducted by any Federal, State, or local authority.” Id. § 2.13. The regulatory mosaic provides safeguards attendant to the disclosure and use of substance abuse treatment records in connection with a criminal case. See id. §§ 2.65-2.67. To investigate or prosecute either a substance abuse “program” or a “person holding the records,” law enforcement personnel must obtain a court order premised upon a showing of good cause, id. § 2.66(c) (cross-referencing § 2.64(d)-(e)), and must employ certain prophylactic measures to shield patients’ identities, id. § 2.66(d). While a court may issue a disclosure order without any particular form of prior notice, the regulations require that “upon implementation of an order” the program, the person holding the records, and any patient whose records are disclosed — a group whom we shall call “the protected parties” — “must be afforded an opportunity to seek revocation or amendment of that order.” Id. § 2.66(b). The district court found that the defendant (as the sponsor of CAP) qualified as a “program” for purposes of this regulation and, thus, was entitled to seek revocation or amendment of the ex parte disclosure orders issued by the magistrate judge. 2 Shinderman, 432 F.Supp.2d at 154. Building on this foundation, the court ruled that the “opportunity” to revoke or amend each order arose immediately upon disclosure of the records in question to law enforcement personnel, that the regulations mandated immediate notice to the protected parties at that juncture, and that the window of delay countenanced by the magistrate judge violated section 2.66(b). Id. The court concluded, however, that since neither bad faith nor prejudice to the defendant existed, suppression “was too drastic a remedy.” Id. *12 Before us, the defendant claims that once the district court found a regulatory violation, another section of the rules — 42 C.F.R. § 2.13 — required the suppression of any evidence disclosed pursuant to the tainted order. His claim raises potentially important questions anent the meaning of the regulations and the appropriate use of the exclusionary rule to address a regulatory infraction. When all is said and done, however, we need not reach those questions; after all, “[w]e are not wedded to the lower court’s rationale, but, rather, may affirm its order on any independent ground made manifest by the record.” InterGen N.V. v. Grina, 344 F.3d 134, 141 (1st Cir.2003). In this case, any application of the exclusionary rule must be premised on an underlying regulatory violation — and we discern none. The regulation at issue, 42 C.F.R. § 2.66(b), is entitled “[n]otice not required.” Its principal purpose is to invest the court with discretion to grant disclosure orders ex parte, that is, without notice to any protected party. It is thus apparent that neither the text nor the purpose of the regulation supports the notion that a delay in notice is per se unlawful. To be sure, the regulation does direct that protected parties be afforded an opportunity to revoke or amend any judicial disclosure order. But in setting forth that directive, the regulation nowhere makes explicit a duty to notify the protected parties at a particular point in time. As we read its plain language, section 2.66(b) requires only that protected parties be given a chance to revoke or amend the order “upon [its] implementation.” It would be perverse to graft an immediate notice requirement onto a regulation that explicitly grants district courts the discretion to relax notification requirements. In our view, section 2.66(b) should be construed as it is written. Its text demands that a court issuing a disclosure order afford protected parties with an opportunity to contest the underlying validity and scope of the disclosure — nothing more. Of course, that opportunity must be meaningful; otherwise, the regulation would hold out only an empty promise. With this qualification in mind and with full realization that each individual case depends largely on context, we are confident that notice of a disclosure order does not always have to be coterminous with the entry of the order. An opportunity to challenge a judicial decree does not have to arise instantaneously in order to be meaningful. Analogies are plentiful. Indeed, it is commonplace for a court to delay notice of a search or seizure in order to protect the integrity of a criminal investigation. See, e.g., 18 U.S.C. § 2518(8)(d) (allowing district court to withhold notice from a person subject to a wiretap for up to ninety days after the termination of the underlying order); id. § 3103a(b) (giving district courts discretion to delay any notice required in connection with the issuance of a search warrant intended to obtain evidence of a criminal offense). As long as the delay in notice does not erode a protected party’s right to challenge the order, we believe that section 2.66(b) allows judges to exercise discretion as to the timing and form of post-order notice. Up to the point where prejudice looms, it makes no difference when the opportunity to seek the revocation or narrowing of a disclosure order arises. This reading of section 2.66(b) does not eviscerate the confidentiality protections afforded by the regulatory scheme. A court can issue an ex parte disclosure order only after a finding of good cause. See 42 C.F.R. § 2.66(c) (cross-referencing *13 § 2.64(d)). Even then, no document released pursuant to such an order can be used to investigate or prosecute a patient. Id. § 2.66(d)(2). In this vein, no public disclosure can be made without redaction of all patient-related information. Id. § 2.66(d)(1). And, finally, all protected parties retain the right to challenge the issuance of the order. Given these features of section 2.66 and its eschewal of pre-order notice, reading that regulation to require notice immediately upon release of the records to law enforcement personnel would seem to create a curious anomaly and, in the bargain, rein in the district court’s normal discretion without advancing any significant interest in confidentialityAgainst this backdrop, we turn to the specifics of the instant case. The record makes manifest that the delay in notice did not erode the defendant’s ability to contest the validity and scope of the court orders. The periods of delay authorized by the magistrate judge were not unreasonable. More importantly, the district court found as a fact, at the hearing on the motion to suppress, that the delay did not prejudice the defendant. Shinderman, 432 F.Supp.2d at 154-55. That finding is un-impugnable: the defendant has identified no argument that he could have employed immediately upon the issuance of a disclosure order that was adversely affected by a delay in notice. 3 In all events, any profession of prejudice would be undone by the defendant’s failure to move under section 2.66(b) for revocation or amendment. After all, once the government notified him of the issuance of the disclosure orders, he had the opportunity to object to their validity and scope— yet he conspicuously failed to move for relief at that point. Instead, he did nothing for almost two years and launched an effort to suppress the fruits of the orders only after the grand jury handed up an indictment. The short of it is that no prejudice occurred here. Given this lack of prejudice, the magistrate judge’s authorization of brief periods of delay in the giving of notice did not violate the defendant’s rights under the regulatory scheme. The lower court’s denial of the defendant’s motion to suppress was, therefore, unexceptionable. B. Entrapment. The defendant next challenges the district court’s refusal to instruct the jury on his envisioned entrapment defense. We review a claim of erroneous failure to instruct a criminal jury on a proffered theory of defense de novo. United States v. Sánchez-Berríos, 424 F.3d 65, 76 (1st Cir.2005); United States v. Rodríguez, 858 F.2d 809, 812 (1st Cir.1988). In conducting that assay, we must refrain from differential factfinding and examine the evidence in the light most favorable to the accused so as to determine whether the record supports an entrapment theory. See United States v. Ramos-Paulino, 488 F.3d 459, 462 (1st Cir.2007); Rodriguez, 858 F.2d at 812. *14 Entrapment is an affirmative defense. See United States v. Gifford, 17 F.3d 462, 468 (1st Cir.1994). To undergird that defense, the accused must make a prima facie showing both that the government induced the commission of the charged crime and that he (the accused) lacked a predisposition to engage in it. 4 See Mathews v. United States, 485 U.S. 58, 62-63, 108 S.Ct. 883, 99 L.Ed.2d 54 (1988); United States v. Luisi, 482 F.3d 43, 52 (1st Cir.2007). This is merely a burden of production; once it is satisfied, the government must prove beyond a reasonable doubt that no entrapment occurred. See Sánchez-Berríos, 424 F.3d at 76; Rodríguez, 858 F.2d at 814-15. The issue here is whether the defendant satisfied his entry-level burden. This burden, though modest, see United States v. Gamache, 156 F.3d 1, 9 (1st Cir.1998), requires more than self-serving assertions. The defendant must adduce “some hard evidence” that “governmental actors induced [him] to perform a criminal act that he was not predisposed to commit.” Rodriguez, 858 F.2d at 814. Here, the district court concluded that the defendant had not cleared this unprepossessing hurdle and, thus, refused to instruct the jury on the entrapment defense. We test the strength of this conclusion. As enunciated at trial, the defendant’s theory of entrapment was configured as follows. He presented evidence which, if credited, was competent to show that he possessed only a temporary Maine medical license; that the DEA improperly withheld the issuance of a Maine-specific DEA registration number; that JoAnn Masar, a DEA agent, informed him that a permanent Maine medical license was necessary in order to receive a Maine-specific DEA number; 5 that this information was incorrect; and that, therefore, he had no choice but to use another physician’s name and DEA number to prescribe the benzodiaze-pine that his patients needed. This evidence is inadequate to satisfy the defendant’s entry-level burden of showing improper inducement. Taking that evidence in the light most flattering to the defendant, the most that can be said is that Masar’s incorrect information and the DEA’s improper withholding of a Maine-specific DEA number were “but for” causes of the later criminal conduct. That is not enough: the tort concept of “but for” causation does not constitute the appropriate test for inducement, and that concept is ultimately too attenuated a link on which to hinge an entrapment defense. Improper inducement requires a showing not only that the government afforded a defendant an opportunity to commit a crime but also that it brought to bear “something more — something akin to excessive pressure, threats, or the exploitation of an unfair advantage.” RamosPaulino, 488 F.3d at 462; see also Sorrells v. United States, 287 U.S. 435, 454, 53 S.Ct. 210, 77 L.Ed. 413 (1932) (Roberts, J., concurring) (defining entrapment as “the conception and planning of an offense by an officer, and his procurement of its commission by one who would not have perpe *15 trated it except for the trickery, persuasion, or fraud of the officer”). In this instance, the defendant’s evidence does not even show that the government furnished him with an opportunity to commit a crime, much less that the government took unfair advantage of him. We cannot tease from the defendant’s proffer anything that might reasonably be expected to encourage an individual to engage in criminality. See, e.g., Gamache, 156 F.3d at 9 (explaining that government action must “ereate[ ] a risk of causing an otherwise unwilling person to commit the crime charged”). Even if the defendant was fed incorrect information and the issuance of a DEA number was wrongfully withheld, there is simply no evidence that Masar ever urged or instructed the defendant, nor even hinted to him, that he should try to evade the law either by using another physician’s DEA number or by forging another physician’s signature. The archetypical entrapment case involves a government “sting” operation, see, e.g., United States v. Gendron, 18 F.3d 955, 961 (1st Cir.1994), or some other situation in which the government procures the defendant’s commission of a crime, see, e.g., Sorrells, 287 U.S. at 454, 53 S.Ct. 210, or at least a situation in which the government provides material assistance to the defendant in order to enable him to commit a crime, see, e.g., United States v. Russell, 411 U.S. 423, 427, 93 S.Ct. 1637, 36 L.Ed.2d 366 (1973). This case does not come close to fitting that mold. At the expense of carting coal to Newcastle, we add that even if the government provided the defendant with the opportunity to commit a crime — which it did not — the necessary “plus” factor is entirely lacking. There is no evidence of excessive pressure, threats, or unfair exploitation on the government’s part. There is no suggestion that the government used coercive tactics, see, e.g., United States v. Stanton, 973 F.2d 608, 610 (8th Cir.1992); employed any hard-sell rhetoric or dogged insistence, see, e.g., Rodriguez, 858 F.2d at 815-16; or engaged in any arm-twisting or exploitation of economic need, see, e.g., Sherman v. United States, 356 U.S. 369, 376, 78 S.Ct. 819, 2 L.Ed.2d 848 (1958). While such evidence, standing alone, would be insufficient to demonstrate improper inducement, it may in certain circumstances be a starting place for mounting a colorable showing of inducement. To be sure, we have suggested that, in certain circumstances, a government agent’s false statements might sustain a finding of improper inducement. See Gamache, 156 F.3d at 9. But in the classic entrapment context, false statements constitute improper inducement only when they have the purpose or effect of preying on the sympathy, weakness, or goodwill of a defendant. See, e.g., United States v. Sullivan, 919 F.2d 1403, 1419 & n. 21 (10th Cir.1990) (telling defendant that undercover agent was suicidal and in desperate need of money). We discern no proof of any such manipulative behavior here. We have said enough on this score. 6 We conclude, without serious question, that the entrapment defense cobbled together here was a mirage and that the district court appropriately declined to instruct the jury on it. C. Cross-Examination. The defendant further complains that the district court violated Rules *16 403 and 608(b) of the Federal Rules of Evidence when it allowed the government to cross-examine him, over objection, about his responses to certain questions when applying for medical licensure in Maine. We review the disputed ruling for abuse of discretion. See, e.g., United States v. Brown, 500 F.3d 48, 58 (1st Cir.2007); United States v. Winchenbach, 197 F.3d 548, 557 (1st Cir.1999). We set the stage. The defendant applied for a Maine medical license in 2001 and again in 2002. As part of each application, he answered “no” to the following question: “have you EVER been charged, summonsed, indicted, arrested or convicted of any criminal offense, including motor vehicle offenses, but not including minor traffic or parking violations?” 7 Evidence available to the government suggested that the defendant had been arrested in 1965 and again in 1970 for drug-related offenses (although neither arrest culminated in a conviction). At trial, the defendant testified in his own behalf. The government sought to cross-examine him about these answers, seeking to cast doubt upon his credibility. The defendant objected and moved in li-mine to exclude any such inquiry. While he admitted that he had been arrested, he asserted that the arrests had been expunged and that, therefore, he had answered the questions truthfully and in line with the advice of counsel. In support of this assertion, he submitted an affidavit from his Illinois lawyer attesting that, to the best of the lawyer’s recollection, the arrests had been expunged. The defendant also argued, in the alternative, that any probative value that might be derived from the cross-examination would pale in comparison to the prejudice that would accrue. The district court concluded that the carefully worded affidavit provided “no convincing ground” to support the defendant’s professed belief that the arrests had vanished and did not have to be revealed on the license applications. Then, in the exercise of its discretion, the court ruled that the government could cross-examine the defendant about his arrest-related answers. Withal, the court precluded the government from introducing the arrest records themselves into evidence. The defendant ascribes error to the former ruling. The touchstone of our analysis is Federal Rule of Evidence 608(b), which preter-mits the use of extrinsic evidence of specific instances of a witness’s conduct for the purpose of bolstering or attacking the witness’s credibility. Despite this general proscription, the rule explicitly grants district courts discretion to allow cross-examination of a witness about specific instances of misconduct as long as those instances are probative of his “character for truthfulness or untruthfulness.” Typically, a trial court’s discretion to determine the scope and extent of cross-examination is broad. See, e.g., Delaware v. Van Arsdall, 475 U.S. 673, 679, 106 S.Ct. 1431, 89 L.Ed.2d 674 (1986); United States v. Beltrán, 761 F.2d 1, 11 (1st Cir.1985). That discretion is nonetheless subject to the overarching need to balance probative worth against prejudicial impact. See United States v. Simonelli, 237 F.3d 19, 23 (1st Cir.2001); Fed.R.Evid. 608(b) advisory committee notes. That balancing function is spelled out in Federal Rule of Evidence 403, which states in pertinent part that relevant evidence “may be excluded if its probative value is substantially out *17 weighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury.” Rule 403 judgments are typically battlefield determinations, and great deference is owed to the trial court’s superior coign of vantage. “Only rarely — and in extraordinarily compelling circumstances — will we, from the vista of a cold appellate record, reverse a district court’s on-the-spot judgment concerning the relative weighing of probative value and unfair effect.” Freeman v. Package Mach. Co., 865 F.2d 1331, 1340 (1st Cir.1988). In this instance, we discern no abuse of discretion. After all, a witness’s willingness to lie to the government in an application for a license is highly probative of his character for truthfulness. See, e.g., United States v. Carlin, 698 F.2d 1133, 1137 (11th Cir.1983); see also United States v. Tse, 375 F.3d 148, 166 (1st Cir.2004) (discussing false statements in employment application). Here, moreover, temporal considerations favored use of the evidence; the defendant’s answers were not remote in time but, rather, were roughly contemporaneous with the criminal conduct with which he had been charged. 8 See Simonelli, 237 F.3d at 23. And, finally, the central factual issue at trial revolved around the defendant’s intent. Hence, his credibility was highly relevant to the outcome of the case. See United States v. Mateos-Sánchez, 864 F.2d 232, 237 (1st Cir.1988). Taking all of these factors into account, the district court’s determination that the defendant’s untruthful answers in his license applications were fair game for cross-examination falls comfortably within the encincture of judicial discretion. This brings us to the question of prejudice. While we agree with the defendant that the revelation of his prior arrests carried with it some potential for untoward effect, we do not believe that effect was so stark as to compel the exclusion of the evidence. We long have recognized that “all evidence is meant to be prejudicial; it is only unfair prejudice which must be avoided.” United States v. Rodríguez-Estrada, 877 F.2d 153, 156 (1st Cir.1989) (emphasis in original). Thus, the court appropriately could discount whatever prejudice flowed from the impeaching effect of the answers. Beyond that, the answers themselves were not particularly inflammatory, and the court did not permit the government to elicit any tawdry details. It is also noteworthy that the court took affirmative steps to minimize any risk of unfair prejudice. For example, it allowed the defendant to tell the jury about the ultimate disposition of the arrests and about his belief that they had been expunged. Furthermore, the court offered to give a limiting instruction; that the defendant eschewed this course does not minimize the value of the court’s offer. Given the totality of the circumstances, we find unconvincing the argument that the court committed a palpable error in judgment in allowing the cross-examination to proceed. The defendant cites our decision in United States v. Tse for the proposition that cross-examining a witness about responses to questions included in an employment application violated Rule 608(b). *18 This argument misreads Tse, in which the court permitted the defendant to cross-examine a government witness regarding his answers in an employment application. 375 F.3d at 166. Tse also upheld the trial court’s discretion to deny the defendant the opportunity to refresh the witness’s recollection by showing him the application, itself excluda-ble as extrinsic evidence under Rule 608(b). See id. From this portion of the opinion, the defendant selectively quotes out of context dicta to the effect that “[u]sing the employment application [to impeach the witness] would be a clear violation of Rule 608(b).” Id. (dictum). The defendant misreads this as suggesting a barrier to any and all use of evidence deemed “extrinsic” under Rule 608(b) to refresh a witness’s recollection. The statement can equally as well be read to articulate the less controversial proposition that introducing such extrinsic evidence for the purpose of proving the witness’s truthfulness (as opposed to using it merely to refresh the witness’s memory) would violate the express prohibition of Rule 608(b). We believe the latter reading is sounder and correctly captures the meaning of the Tse court. For one thing, Rule 612, rather than rule 608(b), governs the use of evidence to refresh a witness’s recollection. That rule never has been construed to require that a writing used to refresh a witness’s recollection must be independently admissible into evidence. The case law holds to the contrary. See, e.g., United States v. Kusek, 844 F.2d 942, 949 (2d Cir.1988); United States v. Scott, 701 F.2d 1340, 1346 (11th Cir.1983). The commentators agree. See, e.g., Jack B. Weinstein & Margaret A. Berger, 4 Weinstein’s Federal Evidence § 612.03[3][b], at 612-14 (2d ed.2007). For another thing, as stated in a leading treatise, “[w]hile extrinsic evidence in the form of documents may not be admitted under Rule 608(b) to impeach, nothing in that rule prohibits the use of documents merely to refresh the recollection of a witness who is the target of impeachment.” 28 Charles Alan Wright & Victor James Gold, Federal Practice & Procedure: Evidence § 6117, at 12 (1993 & Supp.2007) (citing United States v. Chevalier, 1 F.3d 581, 584 n. 2 (7th Cir.1993)). Thus, Tse does not help the defendant here. D. Sentencing. In sentencing the defendant, the district court placed him in criminal history category I, set his base offense level at 10, and applied two-level enhancements for abuse of a position of trust and obstruction of justice, see TJSSG §§ 3B1.3, 3C1.1. The court then departed downward, USSG § 5K2.0, on the basis that the defendant’s actions were outside the heartland of offenses typically occurring under the principal statute of conviction, 21 U.S.C. § 843(a)(3). The court sentenced the defendant to a six-month term of immurement and a two-year term of supervised release. On appeal, the defendant protests only the obstruction of justice enhancement. His protestation targets the district court’s finding that he committed perjury at trial (and, thus, obstructed justice). Because this is a direct challenge to a factual finding made at sentencing, our review is for clear error. See United States v. Gobbi, 471 F.3d 302, 314 (1st Cir.2006); United States v. Akitoye, 923 F.2d 221, 229 (1st Cir.1991). The applicable sentencing guideline requires a two-level increase in a defendant’s offense level if the sentencing court finds that “the defendant willfully obstructed or impeded, or attempted to *19 obstruct or impede the administration of justice with respect to the investigation, prosecution, or sentencing of the instant offense of conviction.” USSG § 3C1.1. A finding of perjury falls within the purview of this guideline. See United States v. Dunnigan, 507 U.S. 87, 92-94, 113 S.Ct. 1111, 122 L.Ed.2d 445 (1993); USSG § 3C1.1, cmt. n. 4(b). Perjury consists of false testimony under oath concerning a matter material to the proceeding, as long as the testimony is given “with the willful intent to provide false testimony, rather than as a result of confusion, mistake, or faulty memory.” Dunnigan, 507 U.S. at 94, 113 S.Ct. 1111 (citing 18 U.S.C. § 1621(1)). To impose an obstruction of justice enhancement premised on perjury, a sentencing court must make an independent finding that the elements of perjury have been satisfied. Id. at 95, 113 S.Ct. 1111; Gobbi, 471 F.3d at 314. The record makes pellucid that the district court followed these teachings. In a presentence order, the court carefully evaluated the defendant’s veracity, identified material inconsistencies in his trial testimony, and concluded that he had prevaricated. In the process, the court expressly found that the fabricated portions of the defendant’s testimony were part and parcel of “a willfully false reconstruction” of what had transpired and could not “be attributed to confusion, mistake, or faulty memory.” In particular, the court found irreconcilable differences between the defendant’s testimony and Keefe’s testimony as to whether Keefe had authorized the defendant to coopt his name and DEA number. 9 This was no garden-variety testimonial conflict about a peripheral matter but, rather, a head-on clash about a central issue in the case. Where, as here, the sentencing judge has presided over the trial, we must allow him reasonable latitude for credibility assessments. See Gobbi, 471 F.3d at 314-15; Akitoye, 923 F.2d at 228; see also United States v. Ruiz, 905 F.2d 499, 508 (1st Cir.1990) (explaining that, insofar as factfinding is concerned, “the sentencing court’s choice among supportable alternatives cannot be clearly erroneous”). The defendant strives to weaken this robust finding by suggesting that there was no direct conflict between his testimony and Keefe’s. This suggestion elevates hairsplitting to an art form. A review of the record leaves no doubt about the direct and inescapable contradiction between these two men concerning the defendant’s use of Keefe’s name and DEA number. Keefe testified unequivocally that he did not give the defendant permission to use his name and DEA number in issuing prescriptions; the defendant testified to the contrary. The district court acted within its proper province in concluding that Keefe’s testimony was truthful and that the defendant’s was not. The defendant offers two last arguments. First, he cautions that in evaluating his trial testimony the district court had to afford him “the benefit of any plausible doubt.” Gobbi 471 F.3d at 314. *20 That is true as far as it goes — but it does not go very far. This cautionary language to which the defendant adverts “does not mandate the resolution of every conflict in testimony in favor of the defendant.” United States v. Rojo-Alvarez, 944 F.2d 959, 969 (1st Cir.1991) (quoting United States v. Wallace, 904 F.2d 603, 605 (11th Cir.1990)). It merely resolves in the defendant’s favor “those conflicts about which the judge, after weighing the evidence, has no firm conviction.” Id. (quoting Wallace, 904 F.2d at 605). The firmness of the sentencing court’s conviction here scarcely can be questioned. Finally, the defendant suggests that upholding an obstruction of justice enhancement may chill a criminal defendant’s exercise of his constitutional right to testify. This suggestion is baseless: sentencing enhancements for obstruction of justice premised on perjurious trial testimony do not give rise to Sixth Amendment concerns. See Dunnigan, 507 U.S. at 96, 113 S.Ct. 1111; Akitoye, 923 F.2d at 228. III. CONCLUSION We need go no further. For aught that appears, the defendant was fairly tried, fairly convicted, and fairly sentenced. Affirmed. Footnotes [1] On "July 20, 2003, the magistrate judge granted the government an additional sixty days within which to furnish notice of the issuance of the April 24 disclosure order. [2] The term “program” includes "an individual or entity ... who holds itself out as providing, and provides, alcohol or drug abuse diagnosis, treatment, or referral for treatment.” 42 C.F.R. § 2.11. We assume arguendo (and neither review nor pass upon) the correctness of the district court's holding that the defendant qualifies under this rubric. [3] The absence of prejudice is unsurprising because section 2.66(b) limits challenges "to the presentation of evidence on the statutory and regulatory criteria for the issuance of the court order.” We can perceive no argument based on those criteria that the defendant might have lost in consequence of the delay. For example, to issue a court order under section 2.66, a court must find that “[o]ther ways of obtaining the information are not available or would not be effective,” 42 C.F.R. § 2.64(d)(1), and that "[t]he public interest and need for the disclosure outweigh the potential injury to the patient, the physician-patient relationship and the treatment services,” id. § 2.64(d)(2). [4] The defendant’s claim is one of classic entrapment. He has expressly disavowed any claim of entrapment by estoppel. See, e.g., United States v. Sousa, 468 F.3d 42, 46 (1st Cir.2006); United States v. Bunnell, 280 F.3d 46, 49 (1st Cir.2002). [5] The government contends that any conversation between Masar and the defendant occurred after the defendant had forged Keefe’s name and DEA number. Given the reasoning through which we resolve the entrapment issue, we need not deal with this chronological dispute. [6] Since we find that the defendant failed to carry his entry-level burden as to the first prong of the entrapment defense, we need not inquire as to whether the record contains any evidence showing a lack of predisposition. [7] We quote this question substantially as it • appears in the transcript and in other filings, disregarding minor linguistic variations. We note that the record on appeal does not contain the original applications (and, thus, does not contain the actual question). [8] The defendant attempts to use the gap between the dates of the unreported arrests and the time of trial as a basis for a finding that temporal considerations cut the other way. This tactic is disingenuous. The relevant issue at trial was whether the defendant had lied on his license applications in 2001 and 2002. Consequently, the relevant temporal span is the time between those answers and the occurrence of the alleged criminal acts. [9] While it would be improper to impose an obstruction of justice enhancement "merely ... because the jury rejects the defendant's explanation of the facts and finds him guilty," Gobbi, 471 F.3d at 314, that verdict nonetheless can provide support for the sentencing court’s independent findings, see United States v. Villarmán-Oviedo, 325 F.3d 1, 16 (1st Cir. 2003) (upholding a finding of perjury in part because the jury “must perforce have determined that [the defendant's] testimony was false”). Here, the court noted that the jury must have chosen to credit Keefe’s account. The jury's view thus supported the district court’s rationale. See id.