United States v. Tobin

[PUBLISH] IN THE UNITED STATES COURT OF APPEALS FOR THE ELEVENTH CIRCUIT FILED ________________________ U.S. COURT OF APPEALS ELEVENTH CIRCUIT Nos. 09-13944, 09-13945, 09-13975, APRIL 12, 2012 09-14009, 09-14012 JOHN LEY ________________________ CLERK D. C. Docket No. 08-00118-CR-ORL-22-KRS UNITED STATES OF AMERICA, Plaintiff-Appellee, versus CHRISTOPHER TOBIN, JUDE LACOUR, AKHIL BARANWAL, GEUNNET CHEBSSI, JAMES PICKENS, Defendants-Appellants. ________________________ Appeals from the United States District Court for the Middle District of Florida _________________________ (April 12, 2012) Before TJOFLAT, MARTIN, and HILL, Circuit Judges. MARTIN, Circuit Judge: This prosecution sought to impose criminal liability for the distribution of more than $85 million worth of controlled substances over the Internet from 2002 to 2005. During those years, Jude LaCour owned and directed a company called Jive Network, with various Internet websites to distribute prescription drugs. The government alleged that the websites allowed customers to order controlled substances without submitting any medical records or any prescriptions. The government charged that Christopher Tobin, James Pickens, and Akhil Baranwal, three medical doctors, approved those orders perfunctorily and that Geunnet Chebssi, a pharmacist, dispensed the drugs. Following a jury trial, these defendants were found guilty of multiple charges. On appeal, all five appellants challenge their convictions. LaCour, Baranwal, and Chebssi also contest their sentences. After careful review of the record and the parties’ briefs, and after having had the benefit of oral argument, we affirm the convictions of the five appellants, as well as the sentences of Baranwal and Chebssi. We vacate LaCour’s sentence and remand for re- sentencing before a different district judge. I. FACTS AND PROCEDURAL HISTORY 2 We recite the facts of this case in the light most favorable to the government. United States v. Augustin, 661 F.3d 1105, 1111 (11th Cir. 2011). We will also briefly describe the procedural history. A. FACTS From 2002 to 2005, Jude LaCour owned and operated a company called Jive Network, which used various Internet websites, including hundreds of affiliate websites, to sell prescription drugs. On these websites, customers were able to select the type, quantity, and dosage of drugs that they wanted. To place an order, a customer needed only to complete a brief online questionnaire regarding his or her medical history. Customers were not required to submit prescriptions or to provide medical records. Jive Network did not otherwise seek to verify the identity of those who placed the orders. Jive Network employed physicians to review the orders. The doctors did not conduct a physical examination of the customers or contact the customers’ primary care physician. The only information that the doctors used to decide whether to approve an order was the online questionnaire. When reviewing customer orders, the doctors did not have the option of changing the type, quantity, or dosage of drugs selected by the customer. Once a doctor approved an order, Jive Network’s computer system generated a prescription that included the 3 doctor’s signature. A pharmacist who worked with Jive Network would then fill the prescription and mail the drug to the customer. From 2002 to 2005, Jive Network sold nearly 5 million Schedule III pills and more than 39 million Schedule IV pills. These Schedule III and IV pills accounted for nearly 80 percent of the drugs sold by Jive Network, and they generated an estimated revenue of more than $85 million over the three-year period. At trial, several witnesses testified that they ordered prescription drugs from the Jive Network websites and that, having become addicted to the substances, they would provide false information about their identity in order to obtain the drugs they wanted. Christopher Tobin, Akhil Baranwal, and James Pickens were three of the medical doctors who reviewed and approved the Internet orders for controlled substances. Geunnet Chebssi was a pharmacist who, in turn, filled the orders. During his time with Jive Network, Tobin approved more than 40,000 orders for controlled substances. These orders included one that Lisa Price placed in June 2003 in the name of her daughter, Krista Price, for phendimetrazine (charged in Count 3), as well as one placed by Terry Richards in October 2003 in the name of her son, Tim Richards, for phentermine (Count 11). Tobin spent as little as six seconds reviewing individual customer orders. 4 Baranwal approved more than 61,000 orders. These included one placed by Mary Trerotola for Adipex-P in June 2004 (charged in Count 14), one placed by Kathy Bachand for phentermine in August 2004 (Count 15), as well as one placed by Lisa Price for phendimetrazine in September 2004 (Count 16). Baranwal spent as little as nine seconds reviewing individual customer orders. Pickens approved more than 40,000 orders. These included an order placed by Jamie McCook for Didrex in November 2004 (charged in Count 21). Pickens spent as little as nineteen seconds reviewing individual customer orders. Chebssi filled more than 21,000 prescriptions. These included one for phentermine for Evan Kopald in October 2004 (charged in Count 18). B. PROCEDURAL HISTORY On May 8, 2008, a grand jury returned a seventy-three-count indictment against the appellants, as well as six other defendants.1 On September 17, 2008, a grand jury returned a fifty-three-count superseding indictment against the same defendants. Under Count 1 of the superseding indictment, all five appellants were charged with conspiracy to distribute Schedule III and Schedule IV controlled substances without valid prescriptions in violation of 21 U.S.C. § 846. Under 1 The six other defendants were Jeffrey LaCour (Jude LaCour’s father), Hudsen Smith, Alexis Roman Torres, Andrew Desonia, Abel Lau, and Margaret McIntosh. These defendants did not proceed to trial, but pleaded guilty. They have no part of this appeal. 5 separate counts, the five appellants were also charged with distribution of Schedule III and Schedule IV controlled substances without valid prescriptions in violation of 21 U.S.C. § 841(a)(1).2 LaCour was also charged under Count 32 with conspiracy to commit money laundering in violation of 18 U.S.C. § 1956(h); under Counts 33–51 with transactional money laundering in violation of 18 U.S.C. § 1957; and under Count 52 with concealment money laundering in violation of 18 U.S.C. § 1956(a)(1)(B)(i). On February 11, 2009, the grand jury returned a fifty- three-count second superseding indictment. This indictment differed from the first superseding indictment in only one respect: in Count 1, which charged the appellants and other defendants with conspiracy to distribute controlled substances without valid prescriptions, the word “willfully” was replaced with the word “intentionally.” On March 31, 2009, the case went to trial. On April 30, 2009, the jury convicted all five remaining defendants, appellants here, of distribution of controlled substances in violation of 21 U.S.C. § 841(a)(1). LaCour and Tobin were convicted of conspiracy to distribute controlled substances in violation of 21 2 Counts 2–31 each involved an individual instance of distribution. LaCour was charged under all counts. Tobin was charged under Counts 3 and 11; Baranwal was charged under Counts 14, 15, and 16; Pickens was charged under Count 21; and Chebssi was charged under Count 18. 6 U.S.C. § 846, but the three other appellants were acquitted on the conspiracy count. Finally, the jury convicted LaCour of all remaining counts with which he was charged: conspiracy to engage in money laundering in violation of 18 U.S.C. § 1956(h), transactional money laundering in violation of 18 U.S.C. § 1957, and concealment money laundering in violation of 18 U.S.C. § 1956(a)(1)(B)(i). About three months later, the District Court sentenced LaCour to 97 months imprisonment; Tobin to 36 months imprisonment; Baranwal to 27 months imprisonment; Pickens to 21 months imprisonment; and Chebssi to 15 months imprisonment. All five appellants timely appealed. II. STANDARDS OF REVIEW The appellants raise a multitude of issues on appeal. We review de novo the following questions: whether a statute is unconstitutionally vague, United States v. Duran, 596 F.3d 1283, 1290 (11th Cir. 2010); whether the rule of lenity is applicable, see United States v. Murrell, 368 F.3d 1283, 1285 (11th Cir. 2004); whether an indictment sufficiently presents the elements of the charged offense, United States v. Dabbs, 134 F.3d 1071, 1079 (11th Cir. 1998); whether the district court misstated the law in its jury instructions, United States v. Deleveaux, 205 F.3d 1292, 1296 (11th Cir. 2000); whether the evidence is sufficient to support a conviction, United States v. Chirino-Alvarez, 615 F.3d 1344, 1346 (11th Cir. 7 2010); whether a prosecutor engaged in misconduct, United States v. Epps, 613 F.3d 1093, 1100 (11th Cir. 2010); and whether a sentence is unconstitutional, United States v. Rozier, 598 F.3d 768, 770 (11th Cir. 2010). In general, we review for abuse of discretion a district court’s grant of a motion in limine, United States v. Harrison, 534 F.3d 1371, 1373 (11th Cir. 2008); a district court’s denial of a motion to subpoena a witness under Federal Rule of Criminal Procedure 17, United States v. Link, 921 F.2d 1523, 1528 (11th Cir. 1991); a district court’s evidentiary ruling during trial to which an objection is timely made, United States v. Baker, 432 F.3d 1189, 1202 (11th Cir. 2005); a district court’s limitation on the scope of cross-examination, United States v. Maxwell, 579 F.3d 1282, 1295 (11th Cir. 2009); and a district court’s refusal to give a requested jury instruction, United States v. Svete, 556 F.3d 1157, 1161 (11th Cir. 2009) (en banc). A district court abuses its discretion, however, if it commits an error of law. United States v. Peter, 310 F.3d 709, 711 (11th Cir. 2002). We review de novo questions of law. See Murrell, 368 F.3d at 1285. We review for abuse of discretion a district court’s decision to deny a motion to continue or delay a trial, United States v. Graham, 643 F.3d 885, 893 (11th Cir. 2011); a district court’s decision to deny a motion for severance, United States v. Schlei, 122 F.3d 944, 983 (11th Cir. 1997); a district court’s decision to 8 deny a motion for mistrial based on remarks by the district court, United States v. Tampas, 493 F.3d 1291, 1303 (11th Cir. 2007); a district court’s decision to deny a motion for mistrial based on the jury’s exposure to extrinsic influence, United States v. Ronda, 455 F.3d 1273, 1296 n.33 (11th Cir. 2006); a district court’s investigation of alleged juror misconduct, United States v. Yonn, 702 F.2d 1341, 1344–45 (11th Cir. 1983); and the substantive reasonableness of a sentence, United States v. Jordan, 582 F.3d 1239, 1249 (11th Cir. 2009). An unpreserved objection to a district court decision, such as an evidentiary ruling or its response to a jury question, is reviewed for plain error. See Baker, 432 F.3d at 1202; see also United States v. Wright, 392 F.3d 1269, 1279–80 (11th Cir. 2004). A district court’s decision to deliver an Allen charge is reviewed only to assess whether the charge had a coercive impact. United States v. Trujillo, 146 F.3d 838, 846 (11th Cir. 1998). A district court’s participation in plea discussions constitutes plain error that we may address sua sponte. United States v. Corbitt, 996 F.2d 1132, 1134 (11th Cir. 1993). III. DISCUSSION A. RULE OF LENITY AND VAGUENESS The appellants first argue the Controlled Substances Act (CSA) is unconstitutionally vague as applied to them and that, even if not, the rule of lenity 9 should be applied to reverse their convictions. The essence of both arguments is that at the time of their actions, the appellants did not have fair notice that their conduct would be the subject of criminal prosecution. The government disagrees with both arguments, but like the appellants, seems to suggest that we should resolve the constitutional question before addressing the issue of statutory construction. The Supreme Court, however, has recently reiterated that federal courts must first consider a question of statutory interpretation before addressing a vagueness challenge. Skilling v. United States, 130 S. Ct. 2896, 2929 (2010). Thus, we will discuss the argument regarding the rule of lenity before turning to the constitutional question. 1. Rule of Lenity The rule of lenity “ensures fair warning” by “resolving ambiguity in a criminal statute as to apply it only to conduct clearly covered.” United States v. Lanier, 520 U.S. 259, 266, 117 S. Ct. 1219, 1225 (1997). The Supreme Court has cautioned, however, that “[t]he simple existence of some statutory ambiguity . . . is not sufficient to warrant application of that rule.” Muscarello v. United States, 524 U.S. 125, 138, 118 S. Ct. 1911, 1920 (1998). Rather, for the rule to apply, there must be some “grievous ambiguity” in the statute—that is, an ambiguity that remains even after all of the tools of statutory interpretation are brought to bear. 10 Id.; accord United States v. Camacho-Ibarquen, 410 F.3d 1307, 1315 (11th Cir. 2005) (“We will apply the rule of lenity only if the provision being construed is still ambiguous after application of normal rules of construction.”). “The first rule in statutory construction is to determine whether the language at issue has a plain and unambiguous meaning with regard to the particular dispute. If the statute’s meaning is plain and unambiguous, there is no need for further inquiry.” United States v. Fisher, 289 F.3d 1329, 1337–38 (11th Cir. 2002) (internal quotation marks and citation omitted). If the language is ambiguous, however, we may look to the relevant legislative history. United States v. Dodge, 597 F.3d 1347, 1352 (11th Cir. 2010). We bear in mind that “[s]tatutory construction . . . is a holistic endeavor,” so a statute must be examined as a whole. Id. (quotation marks omitted). As we stated, all five appellants were convicted of distribution of Schedule III and Schedule IV controlled substances in violation of 21 U.S.C. § 841(a)(1). In addition, LaCour and Tobin were convicted of conspiracy to distribute these narcotics in violation of 21 U.S.C. § 846. LaCour and Tobin do not argue that the conspiracy statute is overly vague or that it should be construed narrowly. The key question, then, is the meaning of Section 841(a)(1). 11 Section 841(a)(1) provides that “[e]xcept as authorized by [the CSA], it shall be unlawful for any person knowingly or intentionally . . . to . . . distribute[] or dispense . . . a controlled substance.” 21 U.S.C. § 841(a)(1). Section 829, in turn, authorizes “practitioner[s]” to dispense Schedule III and Schedule IV substances with a “prescription.” Id. § 829(b). Practitioners who seek to dispense controlled substances must register with the Attorney General. Id. § 822(a)(2). The key statutory terms—“controlled substance,” “dispense,” “distribute,” “practitioner”, and “prescription”—are defined either by statute, see id. § 802(6), (10), (11), (21), or by regulation, see 21 C.F.R. § 1306.04(a). In particular, the regulations promulgated by the Attorney General specify that a “prescription” is one that must be “issued for a legitimate medical purpose by an individual practitioner acting in the usual course of his professional practice.” 21 C.F.R. § 1306.04(a). The appellants argue that Section 841(a)(1) is ambiguous and, in so doing, advance two different ideas. First, they suggest that at the relevant time—from 2002 to 2005—Section 841(a)(1) was ambiguous about whether it applied to distributions over the Internet. Second, they argue, also with respect to that period of time, that Section 841(a)(1) was ambiguous about whether it required a doctor to have an in-person patient visit before providing a prescription over the Internet. 12 Both of these arguments rest on the fact that in 2008, Congress passed the Ryan Haight Online Pharmacy Consumer Protection Act, Pub. L. No. 110-425, 122 Stat. 4820 (2008). That legislation amended the CSA to explicitly prohibit the distribution of controlled substances over the Internet. See 21 U.S.C. §§ 841(h), 829(e)(1). The act also defined the term “valid prescription” in the context of the Internet as one made by “a practitioner who has conducted at least 1 in-person medical evaluation of the patient.” Id. § 829(e)(2)(A)(i). Neither of the appellants’ arguments prevails. First, it is clear that even before the Ryan Haight Act was passed, the CSA criminalized the distribution of controlled substances over the Internet. Indeed, the language of the statute makes no distinction among channels of distribution. See 21 U.S.C. § 841(a)(1). The statute, by its own terms, thus makes it unlawful for a practitioner to distribute a controlled substance without a valid prescription—regardless of the channel of distribution. The appellants find it significant that at the relevant time, the CSA did not specifically mention the Internet. We note, however, that the CSA also did not (and does not) explicitly mention in-person, mail, or phone transactions. See id. But these transactions are covered under the statute. The enactment of the Ryan Haight Act does not compel the conclusion that the CSA had previously been ambiguous about whether it applied to the Internet. 13 We have recognized Congress may amend a statute simply to confirm existing law. United States v. Sepulveda, 115 F.3d 882, 885 n.5 (11th Cir. 1997). “Thus, an amendment to a statute does not necessarily indicate that the unamended statute meant the opposite.” Id. (quotation marks omitted). Congress’s decision to amend the CSA in 2008 is best understood as confirming that under the CSA, it is unlawful to distribute controlled substances without a valid prescription, regardless of whether the prescription is made in person, by mail, by phone, or through the Internet. The appellants’ second argument is more sophisticated, but it also fails. The appellants suggest that because the Ryan Haight Act specifically defines a valid Internet prescription as one that is made following an in-person patient visit, 21 U.S.C. § 829(e)(2)(A)(i), the CSA, prior to its 2008 amendment, was ambiguous as to whether it required an in-person visit. It is true that neither the text of 21 U.S.C. § 841(a)(1) nor the text of 21 C.F.R. § 1306.04 speaks to this specific issue. We therefore take the further step of examining the CSA as a whole, as well as its legislative history. See Dodge, 597 F.3d 1347, 1352. Doing so, we conclude that the appellants’ argument must be rejected insofar as it rests on a fundamental misunderstanding of the structure and operation of the CSA and Congress’s decision to enact the Ryan Haight Act in 2008. 14 The overarching aim of the CSA is to combat drug abuse and to control the legitimate and illegitimate traffic of controlled substances. Gonzales v. Oregon, 546 U.S. 243, 250, 126 S. Ct. 904, 911 (2006). Congress in this respect recognized that practitioners have significant access to controlled substances and that as a result, they have “the greatest opportunity” for diverting drugs to illegitimate use. United States v. Moore, 423 U.S. 122, 135, 96 S. Ct. 335, 342 (1975). Thus, a significant feature of the CSA’s “comprehensive, closed regulatory regime,” Gonzales, 546 U.S. at 250, 126 S. Ct. at 911, is the requirement that practitioners be registered with the Attorney General, see 21 U.S.C. § 822(a)(2). As noted, under the CSA, a practitioner who seeks to dispense controlled substances in accordance with 21 U.S.C. § 829 must be registered with the Attorney General to do so. See 21 U.S.C. § 822(a)(2). Congress was undoubtedly aware that by pulling medical professionals into the statute’s ambit, it could easily “encroach on a state’s traditional authority to regulate medical practices.” Oregon v. Ashcroft, 368 F.3d 1118, 1128 (9th Cir. 2004), aff’d sub nom. Gonzales v. Oregon, 546 U.S. 243, 126 S. Ct. 911 (2006). When Congress enacted the CSA, it thus manifested its intent to leave it to the states to define the applicable standards of professional practice. Indeed, Congress directed the Attorney General to register practitioners who were 15 authorized under state law to dispense controlled substances. Pub. L. No. 91-513, tit. II, § 303(f), 84 Stat. 1236, 1255 (1970) (codified as amended at 21 U.S.C. § 823(f)). Congress also provided that the Attorney General could revoke such registration only if, for instance, the practitioner “had his State license or registration suspended, revoked, or denied by competent State authority.” Id. § 304(a)(3) (codified at 21 U.S.C. § 824(a)(3)). In 1984, Congress amended the CSA to authorize the Attorney General to refuse to register, and to revoke the registration of, a practitioner whose registration would be “inconsistent with the public interest.” Comprehensive Crime Control Act of 1984, Pub. L. No. 98-473, §§ 511, 512, 98 Stat. 1996, 2073 (1984) (amending 21 U.S.C. §§ 823(f), 824(a)). This amendment did expand the scope of the Attorney General’s authority, but Congress did not signal any intent to disregard state standards of medical practice. Indeed, in amending the CSA, Congress observed that “because of a variety of legal, organizational, and resource problems, many states are unable to take effective or prompt action against violating registrants.” S. Rep. No. 98-225, at 223 (1984), reprinted in 1984 U.S.C.C.A.N. 3182, 3448. As a result, “even where there is strong evidence that [practitioners] have significantly abused their authority to dispense controlled substances,” they may “continue to dispense drugs.” Id. 16 Congress’s decision to authorize the Attorney General to refuse to register, or to revoke the registration, of a practitioner even when that practitioner maintains his license or registration under state law thus does not reflect an intent to disregard state law. Far from it, Congress was concerned that resource and other constraints limited the ability of the states to enforce their standards of professional practice. See id. The 1984 amendment to the CSA was thus intended to bolster those state standards. The amendment embodies this intent by requiring the Attorney General to consider a practitioner’s compliance with applicable state law, as well as the recommendation of the relevant state licensing board, in determining whether that practitioner’s registration would be “inconsistent with the public interest.” See 21 U.S.C. §§ 823(f), 824(a)(4); see also S. Rep. No. 98- 225, at 223, reprinted in 1984 U.S.C.C.A.N. at 3449 (providing that the Attorney General should “continue to give deference to the opinions of state licensing authorities”). In light of this legislative scheme, which underscores Congress’s desire to defer to the standards of professional practice set by the states, it is not surprising that when the Supreme Court examined the CSA’s structure and operation in Gonzales v. Oregon, it observed that “the statute manifests no intent [on the part of Congress] to regulate the practice of medicine generally.” 546 U.S. at 270, 126 S. 17 Ct. at 923. The Court explained that this was “understandable” because under our federal system, the “regulation of health and safety is primarily . . . a matter of local concern.” Id. at 270, 271, 126 S. Ct. at 923 (quotation marks omitted). The Court acknowledged that Congress has the authority to set “national standards” of medical practice. Id. at 271, 126 S. Ct. at 923. But the Court concluded that there was “only one area in which Congress [had] set general, uniform standards”—namely, the treatment of narcotic addiction. Id. at 271, 126 S. Ct. at 923–24. (Tellingly, even when Congress made the decision to establish national standards in that specific area, it noted that it was “concerned about the appropriateness of having federal officials determine the appropriate method of the practice of medicine.” H.R. Rep. No. 91-1444, at 14 (1970), reprinted in 1970 U.S.C.C.A.N. 4566, 4581.) About two years after the Supreme Court decided Gonzales, Congress amended the CSA by enacting the Ryan Haight Act, and when it did, it confirmed the Supreme Court’s understanding of the basic structure and operation of the CSA, which reflect “the background principles of our federal system.” Gonzales, 546 U.S. at 274, 126 S. Ct. at 925. Indeed, at the time that Congress was considering the Ryan Haight Act, the states had divergent approaches to whether they required practitioners to conduct an in-person evaluation of a patient before 18 issuing a prescription over the Internet. The House Report observed that “all states allow medications to be purchased via the Internet, but some states do not specifically require in-person consultations for prescriptions.” H.R. Rep. No. 110- 869, at 17 (2008), reprinted in 2009 U.S.C.C.A.N. 2130, 2133. Taking note of this divergence, Congress expressed its intent to replace, in the specific area of Internet prescriptions, the state standards with a national one. See id. Congress’s decision to enact the Ryan Haight Act thus underscores the fact that prior to the CSA’s amendment in 2008, the statute was not ambiguous as to whether an in-person consultation was required for a prescription over the Internet to be valid. Rather, consistent with the statute’s “recognition of the state regulation of the medical profession,” Gonzales, 546 U.S. at 270, 126 S. Ct. at 923, the CSA incorporated the applicable state standard on this issue. See H.R. Rep. No. 110-869, at 17, reprinted in 2009 U.S.C.C.A.N. at 2133. The Ryan Haight Act simply reflected a conscious choice by Congress to displace the different state standards in favor of a single, national one on the specific question of whether an in-person patient visit is required for an Internet prescription to be valid. See id. The appellants cite to a number of sources that predate the Ryan Haight Act in order to suggest that the CSA was previously ambiguous about whether an in- 19 person patient visit was required in order for a prescription to be valid. For instance, they point to a 2004 Congressional Research Service report, which stated that it is “not necessarily illegal” for a doctor to write a prescription based solely on an online questionnaire. Jody Feder, Cong. Research Serv., Prescription Drug Importation and Internet Sales: A Legal Overview 18 (2004). In view of the overall structure and operation of the CSA, it is clear that the appellants misunderstand the meaning of these statements. The fact that it was “not necessarily illegal” for physicians to prescribe drugs without seeing a patient in person simply reflects the fact that at the time, the CSA incorporated the state standard at issue, and as Congress recognized, the states had different approaches on this subject. See H.R. Rep. No. 110-869, at 17, reprinted in 2009 U.S.C.C.A.N. at 2133. In sum, the appellants’ argument that the rule of lenity should apply cannot prevail. The appellants stress that prior to the Ryan Haight Act, the CSA was ambiguous about whether it proscribed the distribution of controlled substances over the Internet. But the plain language of the CSA as it then existed makes it clear that the channel through which controlled substances are distributed is of no consequence. The appellants also suggest that prior to its 2008 amendment, the CSA was ambiguous as to whether it required an in-person patient visit for a 20 prescription over the Internet to be valid. To the contrary, Congress’s decision to enact the Ryan Haight Act demonstrates that prior to its amendment, the CSA simply incorporated the applicable state standard on the issue.3 The Ryan Haight Act did not remove an ambiguity in the CSA. It displaced state standards in favor of a national one in the area of Internet prescriptions.4 2. Vagueness The appellants also argue that the CSA is unconstitutionally vague as applied to them. “Void for vagueness means that criminal responsibility should not attach where one could not reasonably understand that his contemplated conduct is proscribed.” Duran, 596 F.3d at1290 (quotation marks omitted). A statute is void for vagueness if it fails to “define the criminal offense [1] with sufficient definiteness that ordinary people can understand what conduct is 3 We agree with the appellants that this is not necessarily a case like Moore, where the conduct that was at issue simply did not meet any applicable state standard of practice. See 423 U.S. at 126, 96 S. Ct. at 338 (noting that the defendant conceded that his conduct “was inconsistent with all accepted methods of treating addicts”). Indeed, during the period of time that is relevant to this case, the states had different approaches on the specific issue of whether an in-person patient consultation was required for a prescription to be issued over the Internet. See H.R. Rep. No. 110-869, at 17, reprinted in 2009 U.S.C.C.A.N. at 2133. Notably the appellants here have not made the argument that their behavior was consistent with the state standard that applied to them or that the District Court should have asked the jury to assess their behavior in light of the applicable state standard. 4 To the extent that a distribution of controlled substance over the Internet took place before the effective date of the Ryan Haight Act, the conduct would be prosecuted under Section 841(a)(1). For any distribution that takes place on or after the effective date of the Act, the prosecution would take place under Section 841(h). 21 prohibited and [2] in a manner that does not encourage arbitrary and discriminatory enforcement.” Id. (quotation marks omitted). In light of our analysis set out above, we must also reject the appellants’ argument that the CSA is unconstitutionally vague as applied to them. In effect, the appellants argue that because prior to its 2008 amendment the CSA did not proscribe specific means of distribution, it is unconstitutionally vague as applied to them. However, we point out again that the CSA did not (and does not) specifically refer to in-person, phone, or mail transactions. Under the appellants’ theory, the CSA would also be unconstitutionally vague as applied to cases where controlled substances are distributed in these ways. The appellants cite no case law to support this anomalous result. The appellants also stress that the CSA did not explicitly indicate that an in-person patient visit was required for a prescription over the Internet to be valid. As we have said, however, the CSA merely incorporated the appropriate state standard on this question. The CSA therefore had “sufficient definiteness.” Id. In summary, the appellants’ argument that the CSA is unconstitutionally vague does not carry the day.5 Accord United States v. 5 All five appellants also suggest that the District Court should have dismissed the indictment because it fails to set forth the essential elements of the offense under Section 841(a)(1). More specifically, they argue that the indictment does not contain sufficient allegations to show that they acted outside the usual course of professional practice. This argument has no merit. Congress has specified that “[i]t shall not be necessary for the United States to negative any exemption or exception set forth in [the CSA] in any . . . indictment.” 21 22 Bansal, 663 F.3d 634, 656–57 (3d Cir. 2011) (rejecting similar vagueness challenge); United States v. Birbragher, 603 F.3d 478, 486–89 (8th Cir. 2010) (same). B. STATE OF MIND UNDER SECTION 841(a)(1) The appellants also challenge the way the District Court addressed issues regarding state of mind under Section 841(a)(1). These issues arose in several contexts, including with respect to the government’s motions in limine and the jury instructions. The appellants argue that the District Court should have allowed evidence of their subjective beliefs that they were not violating the law and that they were acting in the course of professional practice. The appellants also contend that the District Court should have instructed the jury to consider their good-faith beliefs. Before going further, it may be useful to again note the legal framework that governs our analysis. As set out above, Section 841(a)(1) provides that “[e]xcept as authorized by [the CSA], it shall be unlawful for any person knowingly or intentionally to . . . distribute[] or dispense . . . a controlled substance.” 21 U.S.C. U.S.C. § 885(a)(1). In United States v. Steele, 147 F.3d 1316 (11th Cir. 1998) (en banc), we recognized that as a result of this statutory directive, an indictment is not insufficient if it does not include the legal assertion that the distribution of controlled substances occurred outside of the “usual course of professional practice.” Id. at 1319–20. From this holding, it follows that neither is an indictment insufficient if it does not contain factual allegations suggesting that a distribution occurred outside of the “usual course of professional practice.” 23 § 841(a)(1). Section 829, in turn, authorizes “practitioner[s]” to dispense Schedule III and Schedule IV substances with a “prescription.” Id. § 829(b). For a “prescription” to be effective, it must be “issued for a legitimate medical purpose by an individual practitioner acting in the usual course of his professional practice.” 21 C.F.R. § 1306.04(a). A review of these different provisions makes it clear that the appellants have lumped together two separate issues regarding state of mind. First, there is the question of the requisite mens rea under Section 841(a)(1) itself—that is, what is the state of mind that must accompany the act of distributing or dispensing a controlled substance? Second, there is the viewpoint from which the jury must determine whether a prescription is issued for a “legitimate medical purpose” and in the “usual course of professional practice.” 21 C.F.R. § 1306.04(a). 1. State of Mind Under Section 841(a)(1) Itself The District Court dealt with the first question in various contexts, and its treatment of the issue was consistent throughout the entire case. First, in granting one of the government’s motions in limine, the District Court held that the mens rea under Section 841(a)(1) is “knowing and intentional” and that it was not necessary for the government to prove willfulness. The District Court explained that as a result, evidence that the defendants had a sincere belief that their 24 distribution of controlled substances conformed with the law would be “irrelevant” and therefore not admissible as to the distribution counts. Second, Tobin filed a motion under Federal Rule of Criminal Procedure 17 to subpoena Robert Bigelow. The District Court understood that Bigelow, an attorney, would have testified that he informed Tobin that his conduct “was a regulatory matter and did not subject Tobin to criminal charges.” The District Court concluded that this testimony would not be relevant and therefore denied the motion. Third, Tobin requested that the District Court instruct the jury that “[g]ood faith is a complete defense to the charges in the indictment” and that good faith reliance on advice of counsel precludes a finding of guilt. The District Court declined to give this instruction. Fourth, Chebssi also requested that the District Court provide a “theory of defense” instruction. That instruction states, among other things, that “she, at all times, . . . ha[d] no intention to do what the law forbids. It is her position that her subjective intent was, at all times, . . . to comply with the law to the best of her ability . . . .” The District Court also declined to give this instruction. 25 The District Court’s conclusion that the mens rea under Section 841(a)(1) is knowledge rather than willfulness is correct.6 Again, Section 841(a)(1) provides that unless otherwise authorized, “it shall be unlawful for any person knowingly or intentionally to . . . distribute[] or dispense . . . a controlled substance.” 21 U.S.C. 841(a)(1). The language of the statute does not refer, in any way, to willfulness, and as a consequence, we have said that the government only needs to prove that the defendant acted knowingly. See, e.g., United States v. Cruz-Valdez, 773 F.2d 1541, 1544 (11th Cir. 1985) (en banc); see also United States v. Faust, 456 F.3d 1342, 1345 (11th Cir. 2006). Indeed, a “statute’s inclusion of the word 6 “As a general matter, when used in the criminal context, a ‘willful’ act is one undertaken with a ‘bad purpose.’” Bryan v. United States, 524 U.S. 184, 191, 118 S. Ct. 1939, 1945 (1998). In a case involving willfulness, the government must prove that “the defendant acted with knowledge that his conduct was unlawful.” Id. (quoting Ratzlaf v. United States, 510 U.S. 135, 137, 114 S. Ct. 655, 657 (1994)); see also United States v. Dominguez, 661 F.3d 1051, 1072 (11th Cir. 2011); United States v. Haun, 494 F.3d 1006, 1009 (11th Cir. 2007). By contrast, the term “knowingly” means that “the act was performed voluntarily and intentionally, and not because of a mistake or accident.” United States v. Woodruff, 296 F.3d 1041, 1047 (11th Cir. 2002). This state of mind “merely requires proof of knowledge of the facts that constitute the offense,” not knowledge of the unlawfulness of the action. Bryan, 524 U.S. at 193, 119 S. Ct. at 1946. The Supreme Court has recognized that under some statutes, “willfulness” may entail something more than just a bad purpose or the general awareness that the action is unlawful. Specifically, in the context of “highly technical statutes that present[] the danger of ensnaring individuals engaged in apparently innocent conduct”—most notably, the Internal Revenue Code—the government must show that “the defendant was aware of the specific provision . . . that he was charged with violating.” Id. at 194, 118 S. Ct. at 1946–47 (citing Cheek v. United States, 498 U.S. 192, 201, 111 S. Ct. 604, 610 (1991)) (emphasis added). The appellants have not suggested that this form of willfulness is required under the CSA, and we do not see the CSA as the sort of technical statute that requires the government to prove that “the defendant was aware of the duty at issue.” Cheek, 498 U.S. at 202, 111 S. Ct. at 611. 26 ‘knowingly’ tilts against any possibility that Congress intended any additional scienter requirement.” United States v. Polar, 369 F.3d 1248, 1252 (11th Cir. 2004). Because Section 841(a)(1) requires knowledge, and not willfulness, the appellants’ challenges to the rulings of the District Court fail. First, for the distribution counts, the District Court correctly granted the government’s motion in limine to exclude evidence that the defendant subjectively believed that they were not committing the offenses. As the District Court explained, because the mens rea under Section 841(a)(1) consists of knowledge rather than willfulness, any evidence that the defendants had a sincere belief that their distribution of controlled substances was in conformity with the law would be “irrelevant” to the distribution counts and thus inadmissible in that respect. Second, for the same reason, the District Court was correct in denying Tobin’s motion to subpoena Bigelow, insofar as this related to the distribution counts.7 Bigelow would have testified that he told Tobin that his conduct would have regulatory, but not criminal, implications. To obtain a conviction of Tobin 7 A district court may deny a Rule 17 motion for different reasons, such as the untimeliness of the request or the failure to show indigency. See Link, 921 F.2d at 1528. The District Court’s decision to deny the Rule 17 motion in this case was based solely on the relevance of Bigelow’s proffered testimony. 27 for the distribution counts, however, the government was not required to prove that Tobin had a “bad purpose either to disobey or disregard the law.” United States v. Haun, 494 F.3d 1006, 1010 (11th Cir. 2007). Bigelow’s testimony would therefore have been irrelevant to the distribution counts. Third, because it was not necessary for the government to prove willfulness under Section 841(a)(1), the District Court properly declined to give Tobin’s proposed “theory of defense” jury instructions as to the distribution counts. Specifically, we have recognized that good faith reliance on counsel is a defense only for crimes that require willfulness. See, e.g., United States v. Langston, 590 F.3d 1226, 1235 (11th Cir. 2009) (noting that the defense is “designed to refute the government’s proof that the defendant intended to commit the offense”); United States v. Brown, 983 F.2d 201, 203 (11th Cir. 1993) (recognizing that advice of counsel may be used to “negate the willfulness element”). Fourth, as we noted, Chebssi also requested that the District Court provide a “theory of defense” instruction. That proposed instruction stated, among other things, that “she, at all times, . . . ha[d] no intention to do what the law forbids. It is her position that her subjective intent was, at all times, . . . to comply with the law to the best of her ability.” Again, this instruction assumes that Section 28 841(a)(1) involves a mens rea of willfulness, and the District Court was correct in rejecting it. 2. State of Mind Under 21 C.F.R. § 1306.04(a) The District Court was not as consistent in its rulings on the question of state of mind under 21 C.F.R. § 1306.04(a). In its order granting one of the government’s motions in limine, the District Court stated that it is “not prohibiting evidence or argument regarding defendants’ good faith with respect to whether they prescribed controlled substances in good faith as part of his/her medical treatment for a patient in the usual course of professional conduct.” However, in its order granting the government’s other motion in limine, the District Court agreed with the government that “the ‘usual course of professional practice’ is [not] a subjective standard.” At the same time, it reiterated that “defendants may offer subjective, good faith evidence of meeting the patient’s needs.” Ultimately, the District Court instructed the jury to consider the question of good faith in determining whether the prescriptions were valid. The District Court said: A controlled substance is prescribed by a physician in the usual course of professional practice and, therefore, lawfully if he or she prescribed the controlled substance in good faith as part of his or her medical treatment for the patient in accordance with the standards of medical practice generally recognized and accepted in the United States. . . . 29 Similarly, . . . a pharmacist is authorized to distribute a controlled substance pursuant to a prescription issued for a legitimate medical purpose by a doctor acting in the course of professional practice. The defendant pharmacist in this case maintains at all times she acted in good faith and in accordance with the standard of pharmaceutical practice generally recognized and accepted in the United States in dispensing medication.8 The seemingly unsteady course of the District Court no doubt reflects the fact that our precedent has not always been clear in specifying the standpoint from which a jury is to determine whether a prescription was “issued for a legitimate medical purpose by an individual practitioner acting in the usual course of his professional practice.” 21 C.F.R. § 1306.04(a). In United States v. Williams, 445 F.3d 1302 (11th Cir. 2006), abrogated on other grounds by United States v. Lewis, 492 F.3d 1219, 1220 (11th Cir. 2007) (en banc), we emphasized that “a physician’s prescribing behavior” must be judged by an “objective standard.” 445 F.3d at 1309. Yet, in the same case, we approved a charge that instructed the jury to consider whether the defendant had a “good faith” belief that he was prescribing medicine in “the usual course of a professional practice.” Id. 8 To the extent that the instruction speaks to “standards of [professional] practice generally recognized and accepted in the United States,” and thus suggests that the CSA imposes national standards of professional practice, it is inconsistent with the Supreme Court’s observation that unless Congress specifies otherwise, the CSA does not establish national standards. Gonzales, 546 U.S. at 269–72, 126 S. Ct. at 922–24. Again, none of the appellants have argued that the District Court should have instructed the jury to assess their behavior in light of the state standards that applied to them. See supra note 3. 30 Similarly, in United States v. Merrill, 513 F.3d 1293 (11th Cir. 2008), we stressed that “[t]he appropriate focus is not on the subjective intent of the doctor.” Id. at 1306. Again, however, we approved a charge that instructed the jury to consider whether the defendant had a “good faith” belief that he was prescribing a controlled substance “in the usual course of professional [practice].” Id. Perhaps muddying the water even further, our decision in Williams cited approvingly to a decision of the Fifth Circuit, which we described as holding that the CSA “provide[s] both subjective and objective measures of the prescribing behavior.” Williams, 445 F.3d at 1310 (citing United States v. Norris, 780 F.2d 1207, 1209 (5th Cir. 1986)). Still, we read our decisions to form a coherent whole. As we have said, the CSA authorizes the distribution of controlled substances by a practitioner so long as the prescription is “issued for a legitimate medical purpose [and] in the usual course of [the practitioner’s] professional practice.” 21 C.F.R. § 1306.04(a). Because the CSA prohibits the distribution of prescription drugs that is not authorized, see 21 U.S.C. § 841(a)(1), a distribution is unlawful if 1) the prescription was not for a “legitimate medical purpose” or 2) the prescription was 31 not made in the “usual course of professional practice.”9 In Norris, the Fifth Circuit upheld a charge that instructed the jury to consider 1) whether the doctor subjectively believed that the prescription was for a “legitimate medical purpose” and 2) whether, from an objective standpoint, the controlled substances were dispensed in the “usual course of professional practice.” 780 F.2d at 1209. Our decisions in Williams and Merrill follow this framework. In both cases, the defendants argued that whether a prescription is made in the “usual course of professional practice” must be evaluated from a subjective point of view. See Williams, 445 F.3d at 1309; see also Merrill, 513 F.3d at 1305. In Williams, we rejected this argument, citing with approval the Fifth Circuit’s decision in Norris. Williams, 445 F.3d at 1309–10. In turn, in Merrill, we adhered to our decision in Williams. See Merrill, 513 F.3d at 1306. These cases thus stand for the proposition that a jury must determine from an objective standpoint whether a prescription is made in the “usual course of professional practice.” Our decisions in Williams and Merrill to uphold references to “good faith” reflect our alternative holding that even if a subjective standard were to apply, the jury charges in those 9 In other words, even if a physician is shown to have subjectively believed that the prescription was for a legitimate medical purpose, the prescription may still be considered invalid if it was not made in the usual course of professional practice. See United States v. Nelson, 383 F.3d 1227, 1233 (10th Cir. 2004); see also United States v. Armstrong, 550 F.3d 382, 397 (5th Cir. 2008), overruled on other grounds by United States v. Balleza, 613 F.3d 432, 433 n.1 (5th Cir. 2010). 32 cases were adequate. See Williams, 445 F.3d at 1310; see also Merrill, 513 F.3d at 1306.10 Considering this, the District Court did not commit error. First, the appellants complain that in granting one of the government’s motions in limine, the District Court rejected the notion that a defendant’s subjective belief that he is acting in the “usual course of professional practice” is relevant. The District Court’s conclusion, however, is consistent with our decisions in Williams and Merrill. See Williams, 445 F.3d at 1309; see also Merrill, 513 F.3d at 1305. Second, the appellants suggest that as the trial progressed, the District Court became increasingly reluctant to admit evidence of good faith. To the extent that this is so, its decision, again, was entirely consistent with the holdings of Williams and Merrill, which indicate that whether a prescription is made in the usual course of professional practice is to be determined from an objective, and not subjective, viewpoint. See Williams, 445 F.3d at 1309; see also Merrill, 513 F.3d at 1305. Third, to the extent that the appellants requested that the District Court instruct the jury to consider the defendants’ subjective beliefs that they were acting 10 Chebssi suggests that requiring an objective determination as to whether a practitioner acted outside of the usual course of professional practice will create a “strict liability offense.” This is not a valid concern. The possibility that a practitioner will unknowingly run afoul of the CSA is extremely low. In general, the CSA incorporates the applicable state standard of professional practice, and thus it holds practitioners to standards to which they are already bound. 33 in the usual course of professional practice, the District Court had grounds to reject these requests because they did not provide “a correct statement of the law.” Jordan, 582 F.3d at 1247. Even if we were to accept that the requests did accurately reflect the law, we note that the District Court’s jury instruction did allow the jury to consider the defendants’ subjective beliefs. The District Court’s own charge thus “substantially covered” the appellants’ proposed instructions. Id. at 1248.11 In fact, the District Court’s instruction on the issue of good faith is essentially the same as those upheld in Williams, see 445 F.3d at 1309, and Merrill, see 513 F.3d at 1306. This being the case, we cannot say that the District Court erred in any way. C. STATE OF MIND UNDER SECTION 846 LaCour and Tobin were both convicted of Count 1, which charged them with conspiracy to distribute controlled substances in violation of 21 U.S.C. § 846. Like the issue of the state of mind under Section 841(a)(1), the question of the state of mind under Section 846 came up before and during trial. First, in ruling 11 Chebssi argues that the District Court should have further defined the term “good faith.” Given that Chebssi was not entitled to an instruction that even referred to good faith, this argument fails. Baranwal separately complains that while the District Court included a reference to good faith in the conspiracy count, it did not do so with the substantive counts. This Court “determines whether the charges as a whole sufficiently instructed the jury.” United States v. Hooshmand, 931 F.2d 725, 731 (11th Cir. 1991) (emphasis added). This jury could have easily referred back to the District Court’s instruction on the conspiracy count to refresh its understanding of the law that applies to the substantive count. 34 on one of the government’s motions in limine, the District Court held that the offense of conspiracy under Section 846 does not require proof of willfulness. The District Court thus concluded that with respect to the conspiracy count, it was appropriate to exclude evidence that the defendants held a good-faith belief that their actions were lawful. Second, the District Court denied Tobin’s Rule 17 motion to subpoena Bigelow. As we have said, the District Court understood that “Bigelow would testify that he told Tobin that [his] conduct was a regulatory matter and did not subject Tobin to criminal charges.” The District Court concluded that “Tobin’s belief that his conduct was not criminally unlawful” was irrelevant, even as to the conspiracy count. Third and finally, the District Court declined to give Tobin’s proposed jury instruction on advice of counsel as it related to the conspiracy count. The proposed instruction stated, in part, that “[g]ood faith is a complete defense to the charge in the indictment” and that Tobin “would not be acting with ‘intent’ to do wrong if, before taking any action with regard to the alleged offense, [he] consulted in good faith [with] an attorney.” LaCour and Tobin argue that these decisions are incorrect and that, as a consequence, their convictions under Count 1 must be reversed.12 12 LaCour and Tobin also argue that the District Court erred when “it allowed the Second Superseding Indictment to delete the term ‘willfully’” from Count 1 of the indictment. This contention lacks merit. The Second Superseding Indictment specifically refers to 21 U.S.C. § 35 1. Exclusion of Evidence As noted, in granting one of the government’s motions in limine, the District Court held that the offense of conspiracy under Section 846 does not require proof of willfulness. In support of this conclusion, the District Court cited the unpublished decision of this Court in United States v. Morales De Carty, 300 F. App’x 820, 828 (11th Cir. 2008). We cannot reconcile that decision, however, with our binding precedent. We have repeatedly recognized that a conviction under Section 846 requires evidence of willfulness on the part of the defendant. See, e.g., United States v. Ruiz, 59 F.3d 1151, 1152, 1154 (11th Cir. 1995) (noting that “willfully” is an element of the conspiracy offense under Section 846).13 The 846. We have previously held that an indictment is sufficient if it “specifically refers to the statute on which the charged [is] based.” United States v. Fern, 155 F.3d 1318, 1325 (11th Cir. 1998). LaCour and Tobin have not otherwise argued that the indictment alleged insufficient facts for the grand jury to find probable cause. See id. 13 See also United States v. Westry, 524 F.3d 1198, 1212 (11th Cir. 2008) (noting that the government must prove that the defendant “knew the essential objects of the conspiracy, which are either to do an unlawful act or a lawful act by unlawful means”); United States v. Frink, 912 F.2d 1413, 1414, 1417 (11th Cir. 1990) (affirming conviction under Section 846 because evidence was sufficient to show that the defendant “had knowledge of the conspiracy’s unlawful purpose”); United States v. Hirst, 668 F.2d 1180, 1184 (11th Cir. 1982) (upholding jury instruction that required the jury to find that the defendant “willfully became a member of [a drug] conspiracy” under Section 846); United States v. Cardona, 650 F.2d 54, 57 (5th Cir. 1981) (noting that to be convicted under Section 846, a defendant must have entered into an agreement with “the purpose of achieving an unlawful objective”); United States v. DeLucca, 630 F.2d 294, 300 (5th Cir. 1980) (noting that “[t]he essence of a conspiracy is the agreement to act in concert for an illegal purpose” and that as a result, the defendant must have had “the purpose of achieving an illegal objective”). Our decisions are in accord with those of other Circuits. See, e.g., United States v. Layne, 192 F.3d 556, 567 (6th Cir. 1999) (“In a § 846 conspiracy, the government must show the 36 crime of drug conspiracy under Section 846 includes the element that the defendant entered into an agreement with “the purpose of achieving an unlawful objective.” United States v. Cardona, 650 F.2d 54, 57 (5th Cir. 1981).14 The District Court appears to have realized this as well. In its final charge, the District Court told the jury that under Section 846, the defendants must have joined the agreement “knowing the unlawful purpose of the plan.” Cf. Eleventh Circuit Pattern Jury Instructions (Criminal) 566 (2010) (instructing the jury to determine whether the defendant “knew the unlawful purpose of the plan and willfully joined in it”). Apart from relying on the unpublished decision in Morales De Carty, the District Court, in granting the government’s motion in limine, also cited the Supreme Court’s decision in United States v. Feola, 420 U.S. 671, 686–87, 95 S. willful formation of a conspiracy and the willful membership of the defendant in the conspiracy . . . .”); United States v. Russell, 109 F.3d 1503, 1513 (10th Cir. 1997) (noting that a “necessary” element of a Section 846 conspiracy is that “the defendants willfully became members of [a] conspiracy”); United States v. O’Campo, 973 F.2d 1015, 1019–20 (1st Cir. 1992) (noting that under Section 846, “a party must be shown to have entered knowingly, willfully and intentionally into an agreement”); United States v. Clark, 928 F.2d 639, 641–42 (4th Cir. 1991) (holding that a Section 846 conspiracy entails “the defendant’s wilful joinder” in an agreement); United States v. Story, 891 F.2d 988, 992 (2d Cir. 1989) (holding that a “defendant’s willful joining” of an agreement is an essential element under Section 846); United States v. Burroughs, 876 F.2d 366, 368–70 (5th Cir. 1989) (noting that a Section 846 conspiracy requires willfulness). 14 In Bonner v. City of Prichard, 661 F.2d 1206, 1209 (11th Cir. 1981) (en banc), this Court adopted as binding precedent all decisions of the former Fifth Circuit handed down before October 1, 1981. 37 Ct. 1255, 1265 (1975). Separately, the government urges us to rely on United States v. Muncy, 526 F.2d 1261, 1264 (5th Cir. 1976). Both of these cases, however, are inapposite because they address the general conspiracy statute of 18 U.S.C. § 371. See Feola, 420 U.S. at 687, 95 S. Ct. at 1265; Muncy, 526 F.2d at 1264. They do not discuss the drug conspiracy statute at issue here, 21 U.S.C. § 846, and for which there is binding precedent. See, e.g., Westry, 524 F.3d at 1212; Ruiz, 59 F.3d at 1152, 1154; Cardona, 650 F.2d at 57; DeLucca, 630 F.2d at 300. Given that willfulness is an element of a drug conspiracy under Section 846, it is clear that the District Court erred when it granted the government’s motion in limine to exclude evidence of “defendants’ lack of knowledge regarding the illegality of the objectives of the conspiracy, or defendants’ good faith belief about the legality of their conduct.” It is also true that the District Court erred in denying Tobin’s motion to subpoena Bigelow, to the extent that Bigelow’s testimony related to the conspiracy charge against Tobin. The District Court understood that “Bigelow would testify that he told Tobin that [his] conduct was a regulatory matter and did not subject Tobin to criminal charges.” But the District Court concluded that “Tobin’s belief that his conduct was not criminally unlawful” was irrelevant. The District Court did not appear to appreciate the fact 38 that Tobin’s subjective belief about the legality of his actions would be relevant with respect to the charge of conspiracy. However, we are mindful that an incorrect evidentiary ruling does not require the reversal of a conviction if it had “no substantial influence on the outcome” of the case. United States v. Hands, 184 F.3d 1322, 1329 (11th Cir. 1999). Here, LaCour has not identified any evidence that he would have presented at trial in support of his good-faith belief. Neither has he argued that the government failed to put on sufficient evidence of his intent. LaCour therefore has not met his burden of showing how he was harmed by the District Court’s grant of the government’s motion in limine. Also, the District Court’s denial of Tobin’s motion to subpoena Bigelow had no connection with LaCour. He has thus failed to demonstrate that the District Court’s error requires the reversal of his conviction as to Count 1. Based on our review of the record, we also conclude that the District Court’s errors had “no substantial influence” on the jury’s decision to convict Tobin of the conspiracy charge. Id. Tobin testified that in 2001, when he became involved in Internet medicine with a company called E-Scripts,15 he met with an 15 The differences in the spelling of this name in this opinion reflect the different spellings for it in Tobin’s brief and the record. 39 attorney named Tom Goolsby. Tobin told the jury that the meeting with Goolsby occurred at his office and that he showed Goolsby the work he was doing at that time. According to Tobin, Goolsby advised him that as a result of his work, he “might be subject to administrative sanctions,” but that he “would incur no criminal liability whatsoever.” Tobin testified that he became involved with Jive Network starting in February 2002. Tobin acknowledged that his previous work with E-Scripts had lingering effects. The Texas Medical Board found that Tobin’s issuance of a prescription for Viagra through E-Scripts violated its regulations, and in November 2002, Tobin entered into a consent order with that board. Tobin testified that he had retained Bigelow as his attorney in connection with the Texas disciplinary action. According to Tobin, “Bigelow would [have] testif[ied] that he told Tobin that [his] conduct was a regulatory matter and did not subject Tobin to criminal charges.” Tobin acknowledged that following the disciplinary action by the Texas Medical Board, he continued to write Internet prescriptions for Jive Network. Tobin conceded that his doing so was in violation of his agreement with the Texas board. In August 2003, Tobin received a notice of charges filed by the North Carolina Medical Board, which also related to his work with E-Scripts. Tobin 40 stated that he continued to write Internet prescriptions after receiving that notice. Notably, Goolsby was called as rebuttal witness by the government. He testified that he met with Tobin once, but that the meeting was at his office, not that of Tobin. He also denied telling Tobin that Tobin’s conduct would not give rise to criminal liability. We agree with Tobin that Bigelow’s testimony, as well as his own testimony about his consultation with Bigelow, would have been “relevant and material” to his case. Indeed, that testimony would have corroborated Tobin’s purported belief that his actions would not lead to criminal prosecution. Nonetheless, we reject the notion that admitting that testimony would have had a “substantial influence” on the ultimate outcome of the case. Hands, 184 F.3d at 1329. “The testimony of a criminal defendant at his own trial is unique and inherently significant.” Nichols v. Butler, 953 F.2d 1550, 1553 (11th Cir. 1992) (en banc). This is particularly true when the subject of the defendant’s testimony is his own state of mind. See United States v. Brown, 53 F.3d 312, 315 (11th Cir. 1995). Indeed, “in common experience circumstantial evidence is most likely to be the only evidence of a subjective state of mind.” United States v. Smith, 548 F.2d 545, 549–50 (5th Cir. 1977) (quotation marks omitted). A defendant’s 41 decision to take the stand and testify to his own state of mind provides the rare opportunity for direct evidence to be presented to the jury. But a defendant’s decision to offer testimony on the issue of mens rea can also be fatal to his attempt to exculpate himself. We have long recognized that “a statement by a defendant, if disbelieved by the jury, may be considered substantive evidence of the defendant’s guilt.” Brown, 53 F.3d at 314. In fact, where there is “some corroborative evidence” of guilt, a defendant’s testimony “may establish, by itself, [the] elements of the offense.” Id. at 314–15 (emphasis added). “This rule applies with special force,” we have stressed, where the element that is at issue is “the defendant’s intent.” Id. at 315. For these reasons, a defendant’s decision to testify in his own defense has “tremendous strategic importance.” United States v. Teague, 953 F.2d 1525, 1532 (11th Cir. 1992) (en banc). Here, Tobin “elected to take the stand and to testify in his defense,” and “hearing [his] words and seeing his demeanor,” Brown, 53 F.3d at 314 (emphasis omitted), the jury apparently rejected Tobin’s testimony that he did not believe that his actions had criminal implications. Beyond that, the jury heard from Tobin that, after the entry of the consent order with the Texas Medical Board, he continued to prescribe drugs over the Internet in knowing violation of the order. The jury also heard from Tobin that after he received a notice of charges from the 42 North Carolina Medical Board, he continued to prescribe drugs over the Internet. All of this provided ample evidence for the jury to find the element of willfulness. See id. at 315. Thus, in light of the jury’s apparent decision to reject Tobin’s testimony about his own state of mind, we conclude that the District Court’s errors did not have a substantial influence on the outcome of the case.16 2. Jury Instruction on Advice of Counsel Tobin also argues that the District Court erred in refusing to give his proposed jury instruction on advice of counsel in connection with Count 1. His proposed instruction states, in relevant part, that Good faith is a complete defense to the charge in the indictment since good faith on the part of Defendant Dr. Tobin is inconsistent with the existence of intent which is an essential part of the charge. . . . So, Dr. Tobin would not be acting with “intent” to do wrong if, before taking any action with regard to the alleged offense, [he] consulted in good faith an attorney whom [he] considered competent, made a full and accurate report to that attorney of all material facts of which [he] had the means of knowledge, and then acted strictly in accordance with the advice given by that attorney. 16 This is not a case where the defendant was precluded entirely from testifying as to his own state of mind. See, e.g., United States v. Kelly, 888 F.2d 732, 743–44 (11th Cir. 1989) (holding that exclusion of the defendant’s testimony was reversible error). Rather, this is a case where the defendant did testify as to his own state of mind but was precluded from further elaborating on that subject. Binding precedent has deemed the exclusion of such evidence to be harmless. See United States v. Wellendorf, 574 F.2d 1289, 1290 (5th Cir. 1978); United States v. Edwards, 458 F.2d 875, 884 (5th Cir. 1972). But cf. United States v. Eisenstein, 731 F.2d 1540, 1546 (11th Cir. 1984) (holding that exclusion was not harmless). 43 “To be entitled to a good-faith reliance [on advice of counsel] instruction, a defendant must show that (1) he fully disclosed all material facts to his attorney; and (2) he relied in good faith on advice given by his attorney.” United States v. Condon, 132 F.3d 653, 656 (11th Cir. 1998). A district court may properly decline to give such an instruction “if it lacks evidentiary support.” Id. (quotation marks omitted). We have recognized that “the burden on the defendant to put forth sufficient evidence to support a proposed jury instruction is low.” United States v. Hill, 643 F.3d 807, 851 (11th Cir. 2011). During his trial, Tobin asked for the instruction based on his testimony regarding his consultation with Goolsby. Again, Tobin testified that he met with Goolsby in his office, where he demonstrated “what [he] was doing on the Internet with these prescriptions with eScripts.” According to Tobin, Goolsby “told me that I might be subject to administrative sanctions,” but that “I would incur no criminal liability whatsoever.” The District Court rejected the instruction in part because it was uncertain that there was sufficient evidentiary support. The District Court also believed that the conspiracy charge did not require an intent to achieve an unlawful purpose. We have concluded, as Tobin urges, that the District Court erred in holding that willfulness is not an element of a conspiracy under Section 846. However, the 44 record does not reveal the necessary evidentiary support for the instruction Tobin sought. The government correctly points out that Tobin consulted Goolsby in 2001 when he was involved with E-Scripts. Tobin thus spoke with Goolsby before he became involved with Jive Network in 2002. It is true that at trial, Tobin told the jury that his work with Jive Network involved “by and large the same things that I was doing with eScripts.” Tobin did not testify, however, that he told this to Goolsby or that he otherwise spoke with Goolsby about his work with Jive Network before working there. The evidence thus did not establish that Tobin “fully disclosed all material facts” regarding his involvement with Jive Network to Goolsby. Condon, 132 F.3d at 656 (emphasis added). The District Court did not abuse its discretion in refusing to provide the instruction as to the conspiracy count. See Hill, 643 F.3d at 851; Langston, 590 F.3d at 1235–36. D. THE MEANING OF “PROCEEDS” UNDER 18 U.S.C. § 1956(a)(1)(B)(i) LaCour was convicted of Count 52, which charged him with conducting a financial transaction that involved the “proceeds” of the distribution of controlled substances with the knowledge that the transaction is designed to conceal the “proceeds.” 18 U.S.C. § 1956(a)(1)(B)(i). The District Court instructed the jury that LaCour could be found guilty of this offense if, among other things, “the funds or property involved in the transaction did in fact represent the proceeds of 45 ‘specified unlawful activity’—in this case, the proceeds of the distribution of controlled substances.” The District Court did not otherwise define the term “proceeds.” On appeal, LaCour argues that the Supreme Court’s decision in United States v. Santos, 553 U.S. 507, 128 S. Ct. 2020 (2008), requires that the term be defined as “profits.” LaCour’s argument fails. In Santos, the Supreme Court was asked to interpret the term “proceeds” in the context of illegal gambling. See 553 U.S. at 509, 128 S. Ct. at 2022–23. Although a plurality of the Court concluded that the term invariably means “profits” rather than receipts, see id. at 514, 128 S. Ct. at 2025 (plurality opinion), Justice Stevens reached the narrower conclusion that “[t]he revenue generated by a gambling business that is used to pay the essential expenses of operating that business is not ‘proceeds’ within the meaning of the money laundering statute.” Id. at 528, 128 S. Ct. at 2033 (Stevens, J., concurring in the judgment). In United States v. Demarest, 570 F.3d 1232 (11th Cir. 2009), we recognized that the Supreme Court’s decision in Santos was fragmented and that as a result, its precedential value lies only in the narrow conclusion announced in Justice Stevens’s concurrence. Id. at 1242. Given that the plurality opinion in Santos is not binding, and that this case involves the distribution of controlled 46 substances rather than an illegal gambling business, the District Court did not err in refusing to define the term “proceeds” as “profits.” See id. (noting that Santos is inapplicable because the case did not involve an illegal gambling operation); see also United States v. Jennings, 599 F.3d 1241, 1252 (11th Cir. 2010) (same). E. SUFFICIENCY OF THE EVIDENCE LaCour, Tobin, and Pickens raise different arguments as to the sufficiency of the evidence underlying some of their convictions. “We review the sufficiency of the evidence de novo, viewing the evidence in the light most favorable to the verdict.” Chirino-Alvarez, 615 F.3d at 1346. Thus, all reasonable inferences and credibility determinations are drawn in favor of the government. See id. 1. Tobin: Counts 3 and 11 Tobin was convicted of Count 3 and Count 11, both of which charged him with distribution of controlled substances in violation of 21 U.S.C. § 841(a)(1). Tobin argues that the evidence was not sufficient to sustain his convictions for these counts because FedEx shipping documents related to those counts did not include tracking numbers or other proof of delivery. We reject this argument. The CSA defines the term “distribute” to mean “to deliver.” 21 U.S.C. § 802(11). In turn, the term “deliver” is defined as encompassing “the actual, constructive, or attempted transfer of a controlled substance . . . .” Id. § 802(8) (emphasis added). 47 Here, evidence of Jive Network’s own records showed that the orders that Tobin approved were shipped. This was sufficient evidence for the jury to find that there was an “attempted transfer” of the controlled substance. Id.17 2. Pickens: Count 21 Pickens was convicted of Count 21, which charged him with distribution of a controlled substance in violation of 21 U.S.C. § 841(a)(1). According to Pickens, the uncontroverted evidence showed that he dispensed the drug charged for a legitimate medical purpose and in the usual course of his professional practice. Pickens emphasizes that prior to joining Jive Network, he also used online questionnaires and prescribed medication without having a face-to-face visit with a patient. Pickens suggests that as a result, his conduct as part of Jive Network was consistent with his usual course of professional practice. We are not persuaded. Pickens does not assert that his conduct was consistent with the standards set by the state where he was practicing. Rather, he emphasizes that his conduct prior to and while he was with Jive Network was the same. This is not a reasonable view of the phrase “the usual course of his 17 Tobin also briefly asserts that the evidence was insufficient for the jury to convict him of the conspiracy charge under Count 1. In making this argument, he erroneously assumes that the substantive count requires a mens rea of willfulness. To the extent that he suggests that the evidence was insufficient to show that he willfully joined a conspiracy, we reject this argument as well, for the reasons we set out above. 48 professional practice.” 21 C.F.R. § 1306.04(a). Under Pickens’s theory, if a physician was a drug “pusher” prior to the events for which he was convicted, then he would be absolved of criminal liability for his continued drug pushing. As the Eighth Circuit has stated, “[t]his cannot be the law.” United States v. Smith, 573 F.3d 639, 649 (8th Cir. 2009). If we were to accept Pickens’s argument, we “would allow an individual doctor to define the parameters of his or her practice and effectively shield the practitioner from criminal liability despite the fact that the practitioner may be acting as nothing more than a large-scale pusher.” Id. at 648–49 (quotation marks omitted). 3. LaCour: Counts 32 and 52 LaCour states that he is challenging his convictions for Counts 32 through 52, but his arguments reach only Count 32 and Count 52. First, LaCour suggests that the evidence was insufficient to show that he attempted to “conceal” the funds that he received in connection with Jive Network. The only count that requires an element of concealment is Count 52, which charged LaCour with a violation of 18 U.S.C. § 1956(a)(1)(B)(i). Contrary to LaCour’s assertion, there was evidence showing that the day after he married Tina Morris, he asked her to open a bank account in her name and that during that discussion, LaCour mentioned that the government had seized some of his funds. There was also evidence that the funds 49 deposited into the account that Ms. Morris opened had been withdrawn by LaCour from another account after the government’s execution of a search warrant. LaCour also suggests that a scheme must be complex in order to satisfy the element of concealment. But this is not necessarily so. Indeed, we have “never limited criminal liability under [18 U.S.C. § 1956(a)] to cases involving a long series of transactions.” Blankenship, 382 F.3d at 1130. We have said that “even one transaction can be enough.” Id. The key to a conviction for concealment money laundering is “a purpose . . . to conceal.” United States v. Naranjo, 634 F.3d 1198, 1208 (11th Cir. 2011). A decision to create “complex arrangements” may be helpful to show such an intent. Id. at 1209. But it is not necessary. See id. Here, the evidence that LaCour withdrew the funds after the government had executed a search warrant provided a sufficient basis for the jury to find that LaCour sought to conceal those funds from the government. Thus, the evidence was sufficient to sustain LaCour’s conviction under Count 52. LaCour also argues that the evidence was insufficient to sustain his conviction for Count 32, which charged him with conspiracy to engage in monetary transactions in criminally derived property in violation of 18 U.S.C. § 1957, all in violation of 18 U.S.C. § 1956(h). To obtain a conviction under this count, the government was required to prove that 1) there was an agreement to 50 violate 18 U.S.C. § 1957 and that 2) Jude LaCour knowingly and voluntarily participated in that agreement. See United States v. Kennard, 472 F.3d 851, 856 (11th Cir. 2006). LaCour asserts that there was not enough evidence to demonstrate that there was an agreement between him and his father, Jeffrey LaCour, to violate 18 U.S.C. § 1957 and that his conviction must be set aside in light of our decision in United States v. Johnson, 440 F.3d 1286 (11th Cir. 2006). In Johnson, we found that there was not enough evidence to sustain a conviction under 18 U.S.C. § 1956(h) because there was “no evidence establishing that [the defendant’s alleged co-conspirator] conspired” with him to commit money laundering. Id. at 1295. LaCour’s reliance on Johnson is misplaced. There, we pointedly observed that the government had alleged that the defendant had only one other co- conspirator. Id. at 1294 n.7 (noting that the government did not allege that the defendant had conspired with “other persons known and unknown” to commit money laundering). Under those circumstances, the lack of evidence that the defendant’s alleged co-conspirator had agreed to violate the law was fatal. See id. at 1295–96. Here, in contrast, the government alleged that LaCour agreed not only with his father, but also with “others, both known and unknown,” to violate 18 U.S.C. § 1957. 51 Viewing the evidence in the light most favorable to the government, we conclude that there was sufficient evidence for a reasonable jury to find that there was such an agreement. The evidence showed that Jive Network repeatedly transferred substantial amounts of money to LaCour Medical, a company owned by Jeffrey LaCour, apparently for a weekly “consulting fee” of $25,000. LaCour has not suggested to us that the jury was given an explanation of the nature of the consulting work that was purportedly done by LaCour Medical or of the reasons for the amount that was charged. Thus, the jury was presented with evidence of repeated, and unexplained, transfers of substantial amounts of money from Jive Network to LaCour Medical. The jury could reasonably infer from this evidence that there was an agreement to launder criminally derived funds. See United States v. Seher, 562 F.3d 1344, 1364 (11th Cir. 2009) (noting that an agreement can be proven by circumstantial evidence). F. CONTINUANCE OF THE TRIAL LaCour argues that the District Court erred when it denied his motion to continue the trial at a hearing on March 19, 2009. LaCour contends that the District Court should have granted his request for a continuance because his former attorney was granted leave to withdraw on March 13, 2009. LaCour states that as a result of the District Court’s decision, he did not have enough time to 52 review the evidence and that this substantially prejudiced him. This argument has no merit. A party contesting the denial of a motion to continue must show “specific, substantial prejudice” as a result of the district court’s decision. Graham, 643 F.3d at 893. “[T]he Supreme Court has made it clear that not every denial of a request for a continuance is a denial of due process.” Baker, 432 F.3d at 1248. In particular, if a district court has warned a criminal defendant of the risks of proceeding pro se, and the defendant insists on doing so, it is not an abuse of discretion for the district court to deny the defendant’s later request for a continuance. See Graham, 643 F.3d at 893–95. In this case, LaCour’s former counsel filed a motion to withdraw on March 7, 2009, asserting that the attorney-client relationship had irretrievably broken down. The Magistrate Judge conducted a hearing on March 13, 2009 and “determined that [LaCour’s] decision to discharge counsel has been made knowingly and intelligently, with full awareness of the difficulties and disadvantages of self-representation.” LaCour was also informed that he could seek the appointment of counsel and that the discharge of his former counsel would not provide a basis for a continuance. 53 In light of these warnings, the District Court did not abuse its discretion in denying LaCour’s motion. LaCour was informed of the risks of proceeding pro se, and he has not shown that after he was advised that counsel could be appointed for him, he attempted to have counsel appointed. Thus, LaCour chose to proceed pro se fully advised of the disadvantages that would flow from self-representation. Beyond that, LaCour “does not mention anything specific that [he] would have done differently” at trial had the continuance been granted. Id. at 895. Under these circumstances, the District Court did not abuse its discretion in denying LaCour’s request for a continuance. G. SEVERANCE Baranwal, Pickens, and Chebssi assert that the District Court erred in denying their motions for a severance from LaCour. The three appellants advance two principal arguments on this issue. First, they argue that severance was warranted given that LaCour was separately charged with a large number of money-laundering counts. Second, they stress that they were prejudiced by LaCour’s attempt to defend himself pro se. Neither of these arguments prevail. Federal Rule of Criminal Procedure 14 provides that if the joinder of defendants “appears to prejudice a defendant . . . the court may . . . sever the defendants’ trials.” Fed. R. Crim. P. 14(a). We have interpreted this provision to 54 require a defendant to demonstrate “the lack of a fair trial due to actual, compelling prejudice.” United States v. Chavez, 584 F.3d 1354, 1360 (11th Cir. 2009). The presence of prejudice must be balanced against the interests of judicial economy. Blankenship, 382 F.3d at 1120. 1. Unrelated Evidence We have recognized that severance may be warranted if evidence that is not admissible against one defendant will be admitted against another. Chavez, 584 F.3d at 1360. However, we have rejected the idea that a disparity in the quantity of such evidence requires severance. Schlei, 122 F.3d at 984. To be sure, severance may be appropriate in light of the quality of the evidence that is to be introduced against one defendant. See Blankenship, 382 F.3d at 1123, 1124 & n.25. However, we have indicated that only “gruesome,” “emotional,” or “highly inflammatory” evidence will provide a basis for severance. See id. This is because of the “strong presumption . . . that jurors are able to compartmentalize evidence” in accordance with a district court’s instructions. Id. at 1123. In light of these principles, we must reject the appellants’ argument that their trial should have been severed from that of LaCour because he alone was charged with a large number of money-laundering counts. Again, the mere fact that there may be a significant difference in the quantity of the evidence is not 55 enough to warrant severance. Schlei, 122 F.3d at 984. And the type of evidence that the government introduced in order to prove the money-laundering charges consisted largely of financial records. This is not the sort of “gruesome,” “emotional,” or “highly inflammatory evidence” that would prevent a jury from compartmentalizing the evidence as instructed. Blankenship, 382 F.3d at 1123, 1124 & n.25. The District Court did not abuse its discretion in having the appellants tried together here. 2. Spill-Over Effect from Pro Se Defendant Baranwal, Pickens, and Chebssi also argue that they were prejudiced by LaCour’s decision to proceed pro se. This argument suggests that LaCour’s representation of himself had a “spill over” effect. We have recognized that severance may be warranted “[w]here a cumulative and prejudicial ‘spill over’ effect may prevent the jury from sifting through the evidence to make an individualized determination as to each [d]efendant.” Chavez, 584 F.3d at 1360–61. However, the mere fact that one of the defendants chooses to proceed pro se does not compel the conclusion that severance is warranted. United States v. Knowles, 66 F.3d 1146, 1160 (11th Cir. 1995). And under our precedent, the possibility that a defendant “might have had a better chance of acquittal if he could have gone it alone” is also not sufficient. Chavez, 584 F.3d at 1361. 56 In view of these principles, the appellants’ argument must fail. The District Court took a number of steps to prevent any spill-over effect from taking place. Cf Knowles, 66 F.3d at 1160. For instance, the District Court instructed the jury not to consider as evidence the statements that LaCour made in representing himself. The District Court also instructed the jury to give separate consideration to each defendant. In the end, the jury acquitted Baranwal, Pickens, and Chebssi of the conspiracy count with which they were charged, while convicting them of the various distribution counts. This verdict indicates that the jury was able to “sift through the evidence and make an individualized determination as to each defendant.” Schlei, 122 F.3d at 984 (quotation marks omitted). H. COMMENTS BY THE DISTRICT COURT On April 8, 2009, as LaCour was attempting to cross-examine a government witness and was, as a result, drawing objections from the government, the District Court stated: Let me interrupt for a second. Any information that Mr. LaCour puts in a question or any assertions he makes . . . do not constitute evidence and should not be considered by you as evidence. The only evidence that Mr. LaCour can put in is when he testifies under oath as this witness is doing. Any statements he makes in his capacity as representing himself that appear to be questions or statements of fact do not constitute evidence and you should not consider it as evidence. 57 On April 20, 2009, LaCour began to present his defense. He called as his first witness a former employee of Jive Network, and he sought to elicit testimony regarding the fact that an undercover agent was hired to work at the company. The government objected on the ground that LaCour was trying to show prosecutorial misconduct, and after some back and forth, the District Court told LaCour: The jury does not have before it your claim of prosecutorial misconduct, nor are they to consider it. Please deal with facts that are relevant to the claims made by the government: [i].e., that you set up a process by which doctors and pharmacists filled prescription orders without ever seeing the patient. That’s the problem you’re faced with. Later, outside of the presence of the jury, the appellants moved for a mistrial. The District Court took the motion under advisement. The following morning, the District Court gave a curative instruction: Members of the jury, yesterday I suggested that Jude LaCour . . . deal with facts relevant to the claims made by the government. To the extent you recall my remarks, please understand that you and not myself are the triers of fact in this case. My responsibility is limited to advising you as to the law that applies to the case. I will provide you with an extensive written instruction before final argument begins. Accordingly, to the extent that you remember my suggestions to Jude LaCour, please ignore the same. Later, outside of the presence of the jury, the District Court denied the motions for mistrial in light of this instruction. 58 LaCour, Tobin, Pickens, and Chebssi argue that the District Court’s comment on April 20, 2009 constitutes reversible error. They assert that the comment either represented an incorrect statement of the law or evinced the District Court’s view that LaCour was guilty. LaCour separately contends that the District Court’s comment on April 8, 2009 prejudiced his right not to testify because it gave the impression that he was in fact required to do so. These arguments do not carry the day. A conviction will not be reversed “based upon comments of the trial judge unless the comments are so prejudicial as to amount to denial of a fair trial,” United States v. Ramos, 933 F.2d 968, 973 (11th Cir. 1991), and the appellant shows that the comment had “a clear effect on the jury.” United States v. Morales, 868 F.2d 1562, 1576 (11th Cir. 1989) (quotation marks omitted). A comment does not constitute reversible error if it 1) “occupied but a few seconds of a lengthy trial,” 2) the comment was directed to counsel rather than the jury, and 3) the trial judge advised the jury to disregard the comment. Id. The District Court’s comment on April 20, 2009 that LaCour’s “problem” was that he “set up a process by which doctors and pharmacists filled prescription orders without ever seeing the patient” was improper. However, the District Court provided a curative instruction that reminded the jury that it was the sole trier of 59 fact and that the applicable law would be provided later on in the District Court’s instructions. Beyond this, the trial lasted about a month, and the comment was directed at LaCour, and not the jury, in an effort to explain to him the basic requirement of relevance under the rules of evidence. The District Court’s comment on April 20, 2009 does not constitute reversible error.18 See id. at 1576–77. LaCour’s complaint that the District Court’s comment on April 8, 2009 prejudiced his right not to testify is also unpersuasive. For one thing, it takes the District Court’s comment out of context. As the passage above indicates, the District Court was trying to remind the jury of the distinction between, on the one hand, questions and arguments by counsel and, on the other hand, evidence. This difference was something that may have been difficult for the jury to keep in mind given that LaCour was proceeding pro se. Later, the District Court instructed the jury that the defendants have “an absolute right not to testify” and that guilt may not be inferred from a defendant’s silence. Under these circumstances, the District 18 Pickens cites to other comments that the District Court made while LaCour was attempting to question witnesses, but he does not explain why, in his view, they were prejudicial. In any event, these additional comments were encompassed by the curative instruction. Pickens also points to comments that the District Court made outside of the presence of the jury. Pickens does not argue that the comments demonstrated “pervasive bias” against a defendant. United States v. Ramos, 933 F.2d 968, 973 (11th Cir. 1991). Instead, he appears to say that the District Court misunderstood the law. Because these comments were made outside of the presence of the jury, however, they cannot provide the basis for reversal. Hill, 643 F.3d at 845. 60 Court’s comment does not constitute error. See Tampas, 493 F.3d at 1303; United States v. LaChance, 817 F.2d 1491, 1497 (11th Cir. 1987). I. JURY TAMPERING AND ALLEGED JUROR MISCONDUCT On the morning of April 20, 2009, some of the jurors indicated to the District Court that they had received envelopes that appeared to contain letters. The District Court first met with the jurors, instructing them not to discuss the envelopes or their content while it was deciding what to do. The District Court then met with counsel and LaCour, telling them that about half of the jurors had received letters designed to influence the case. The letter, it turned out, urged the jurors to find the defendants not guilty, arguing that their conduct was not criminal prior to the passage of the Ryan Haight Act. The District Court questioned the impaneled and alternate jurors individually and under oath in the presence of counsel and LaCour. Six of the impaneled jurors said that they either did not receive the letter or that they had not opened it. The remaining six impaneled jurors and the three alternate jurors opened the letter, but quickly stopped reading.19 Each juror testified that he or she 19 The jurors said that they stopped reading “as soon as I saw Jive Network on it,” “a couple of sentences into it,” “immediately” after seeing the name of a defendant, or after he or she “started seeing the first half” or otherwise read “just the top part,” “first two lines,” “first line,” or “first three or four sentences.” 61 did not discuss the content of the letter with any other juror. One juror said that the letter she received had another juror’s name on the envelope. That juror denied sending the letter. The District Court did not allow counsel or LaCour to question the jurors. Following the questioning, the District Court informed counsel and LaCour that it was going to instruct the jurors that they were not to draw any inferences from the letters and that the case was to be decided on the facts and the law. The District Court then instructed the jury that it would be improper for you to speculate as to who might be responsible for [the incident] or to hold that against any one of the defendants on trial. So please put that out of your mind. . . . This case is to be decided by the testimony and the evidence as you find it to be and applying the law as I instruct you on it. Later on, outside of the presence of the jury, counsel for Baranwal asked to see the content of the letter. The District Court advised that because the FBI was looking into the matter, the letter would be made available only after the FBI “gets through figuring out who it is that’s trying to manipulate this jury.” The defendants then expressed concern about the fact that one juror was under the impression that another juror had sent the letter and that the jury was otherwise tainted by the communication. The defendants moved for a mistrial, and the District Court denied the motion. 62 On that same day, the District Court also received a letter. This letter purported to come from an “independent neutral juror.” The writer said “I am one of the jurors in the . . . case,” and “regardless of the facts presented to me and regardless of the instructions of law given to me, my verdict will be not guilty for all defendants on all counts.” The writer cited as his reason the apparent ambiguity of the CSA. The District Court found this letter “not credible” and chose not to disclose it to the parties or take further action. The jurors did not receive this letter. On April 22, 2009, a majority of the jurors turned in a second letter that they had received. This letter purported to come from a “freelance reporter” and urged the jurors to “think twice before giving a verdict” because the CSA was “vague” before its 2008 amendment. That morning, after presentation of the evidence, the District Court told the jury: I appreciate that you are bringing to our attention any attempts that are being made to communicate with you. . . . I don’t want you to attempt to assign blame to any one defendant. It would be inappropriate to do that. . . . I want you to continue to accept the proposition that in no way should you attempt to speculate on or hold any one defendant responsible for the conduct that you have been subjected to. . . . I don’t want you to be influenced by it or attempt to assign responsibility to any one defendant. That would be wrong. You should decide this case based upon the evidence that you have heard here in the case and the instructions that the Court will give you. 63 Later during the day, as the jury instructions were being finalized, the District Court declined to make the second letter available to counsel and LaCour, but summarized its content. In its final jury charge, the District Court told the jury: Remember, this case must be decided only upon the facts as you determine them to be and the law that I instruct you on. Furthermore, once again, I want to emphasize that you should not hold against any defendant in this case as some indication of guilt the fact that . . . you have been the recipient of this correspondence which you have turned over to us which is now under investigation. It would be entirely improper for you to try to figure out who is responsible and to consider that in your deliberations. Please do not do that. To conclude, make your decision based only on the evidence that you saw or heard here in court. At the conclusion of the forfeiture proceedings a few weeks later, the District Court questioned the jurors about the second letter, and none indicated that he or she had opened it or read it. 1. Extrinsic Influence on the Jury All five appellants argue that their Sixth Amendment right to a fair trial was undermined by these events. This argument is not ultimately persuasive. The Sixth Amendment’s guarantee of trial by an impartial jury requires that a jury decide a case based solely on the evidence that is admitted at trial and in accordance with the court’s instructions on the law. United States v. Siegelman, 640 F.3d 1159, 1182 (11th Cir. 2011). “A juror’s exposure to extraneous material 64 or influence requires a new trial if the exposure poses a reasonable possibility of prejudice to the defendant.” United States v. Khanani, 502 F.3d 1281, 1291 (11th Cir. 2007) (quotation marks omitted). The defendant bears the initial burden of showing that the jury has been exposed to extrinsic material. Ronda, 455 F.3d at 1299. “Once the defendant establishes that such exposure in fact occurred, prejudice is presumed and the burden shifts to the government to rebut the presumption.” Id. The government must show that the extraneous material or influence “was harmless to the defendant.” Id. This Court considers “the totality of the circumstances” in determining whether the government has rebutted the presumption. Id. Relevant factors include: 1) the nature of the extrinsic material; 2) the manner in which the information reached the jury; 3) the factual findings of the district court and the manner of the court’s inquiry into the issue; and 4) the strength of the government’s case. Id. at 1299–1300. Considering these factors, we cannot say that the events posed “a reasonable possibility of prejudice” to the appellants. Khanani, 502 F.3d at 1291 (quotation marks omitted). First, the extrinsic material came in the form of letters contained in envelopes. As for the first letter, the jurors said that they either did not receive the letter, did not open it, or stopped reading just a few lines into it. Each juror 65 stated that he or she did not discuss the content of the letter with any other juror. Similarly, with respect to the second letter, no juror indicated that he or she opened it or read it. The jurors thus had little or no exposure to the material. Second, as the government points out, even if the jurors had read the content of either of the letters, this exposure would have been favorable to the defendants. Indeed, the letters urged the jury to acquit the defendants on the ground that their conduct was lawful prior to the passage of the Ryan Haight Act. Third, the District Court was careful to ascertain the degree to which each juror had read the first letter. The District Court questioned all of the jurors individually and under oath in the presence of counsel and LaCour. The District Court then instructed the jury to disregard the letter and the surrounding events and to decide the case based solely on the evidence presented at trial and the applicable law. When the jurors brought to the District Court’s attention the second letter, the District Court once again reminded the jurors of their obligation to decide the case based solely on the evidence adduced at trial and in accordance with the Court’s instructions on the law. The District Court reiterated this instruction in its final charge.20 Fourth and finally, none of the appellants have 20 The District Court was very aware that it was not the letters themselves, but rather the attempt at jury tampering, that posed the risk of influencing the jury’s decision. The District Court repeatedly instructed the jurors not to draw any inference from the events, however, and 66 suggested that the government’s case was so weak that it was reasonably possible for the jury to return guilty verdicts only on account of the events that transpired. The appellants suggest that the District Court handled the letters in a deficient manner. Specifically, they complain that the District Court did not adequately question the jurors about the first letter, did not allow the parties to question the jurors about the first letter, and did not question the jurors about the second letter immediately after it was informed of that communication.21 These arguments are not convincing. To begin, we note that the appellants largely failed to raise these concerns to the District Court when they had the opportunity to do so. For instance, the record reveals that no objections were made when the reminded them of their duty to decide the case based solely on the evidence admitted at trial and the applicable law. We “must presume that the jurors followed this instruction.” United States v. LaSpesa, 956 F.2d 1027, 1033 (11th Cir. 1992). 21 The appellants also raise two other arguments that lack merit. First, they complain that at the very beginning of the day on April 20, 2009, the District Court spoke to the jury without counsel present. The record shows that during what may have been less than a minute, the only matter of substance that the District Court conveyed to the jurors was its instruction that they “not discuss among [themselves] the envelopes or what was in the envelope to [the] extent anything [they] saw was in the envelope.” The District Court added: “that’s the only purpose of [me] being here is to ask you not to discuss it among yourselves.” We cannot say that this was inappropriate or that it prejudiced the defendants. Second, the appellants argue that the District Court erred in not providing the letters to them immediately. As the District Court explained, however, the material had been turned over to the FBI, and copies of the letters would be available only after the FBI was done with the material. Beyond that, the District Court summarized the content of the letters to counsel. And as promised, the District Court entered the letters into record. Having now seen the letters, none of the parties suggest that the District Court’s summary was inaccurate or that they would have acted in any different way had they seen the letters earlier. 67 District Court informed the parties that it would not allow them to question the jurors about the first letter. Nor does the record contain any sign that the parties requested the District Court to question the jurors about the second letter. More to the point, our precedent is “uniform in holding that the district court has broad discretion to choose the inquisitorial tools . . . necessary and appropriate to determine prejudice where juror misconduct or extrinsic influences are alleged.” United States v. LaSpesa, 956 F.2d 1027, 1033 (11th Cir. 1992). “There is no magic formula that the trial court must follow in conducting th[e] inquiry.” United States v. Savage, 701 F.2d 867, 871 (11th Cir. 1983). We have recognized that while the district court must ascertain the extent to which the jurors have been exposed to extraneous material, there is the risk that dwelling on the matter would only serve to underscore or otherwise draw attention to the material or surrounding events. LaSpesa, 956 F.2d at 1033. The District Court’s decision to ask only certain questions with respect to the first letter, and its decision not to allow the parties to question the jurors, do not exceed the bounds of the “broad discretion” it had. Id. Given that the jurors either did not open the first letter they received or only read a few lines of that document, the District Court could have properly concluded that any further questions about the “jurors’ thoughts, beliefs, suspicions, and the like” or any 68 questioning by the parties themselves would have led the jurors to dwell on the matter, all to the detriment of their decision-making process. See id. The District Court also acted within its discretion in handling the second letter. It is true that the judge did not immediately question the jurors about that letter. However, given the District Court’s individual examination of the jurors about the first letter, its repeated admonition that the jury decide the case based solely on the evidence admitted at trial and in accordance with the Court’s instructions, and the jury’s cooperation in bringing the second letter to the District Court’s attention, the District Court could have properly concluded that the jurors did not open or read the second letter and that as a result, it was not necessary to question the jurors immediately about that communication. See United States v. Barshov, 733 F.2d 842, 851–52 (11th Cir. 1984). In summary, our review of the record does not leave us with the impression that the District Court abused its discretion in handling the attempts at jury tampering or that the letters and surrounding events posed a reasonable possibility of prejudice to the appellants. 2. Alleged Juror Misconduct The appellants also argue that the District Court erred in failing to disclose the letter it received. Again, this letter purported to come from an “independent 69 neutral juror.” The writer said that “I am one of the jurors in the . . . case” and that “regardless of the facts presented to me and regardless of the instructions of law given to me, my verdict will be not guilty for all defendants on all counts.” The writer cited as his reason the apparent ambiguity of the CSA prior to its 2008 amendment. The District Court did not question the jurors about this letter and chose to take no further action. We have recognized that “[a]ny challenge to the district court’s investigation [of alleged juror misconduct] must be viewed in the context of the broad discretion afforded a trial judge confronted with such an allegation.” Yonn, 702 F.2d at 1344. The district court’s discretion “extends even to the initial decision of whether to interrogate the jurors.” Id. at 1345. We have observed that the district court’s discretion “is at its zenith when the alleged misconduct relates to statements [allegedly] made by the jurors themselves.” United States v. Bradley, 644 F.3d 1213, 1277 (11th Cir. 2011) (quotation marks omitted). The District Court in this case explained that it did not question the jurors about the letter or take further action because it found the letter “not credible.” We are not in a position to reject this determination. By the time the District Court received the letter, it had the opportunity to observe the jurors through voir dire and several weeks of trial. “[T]he district court is uniquely situated to make the 70 credibility determinations” regarding juror misconduct. United States v. Abbell, 271 F.3d 1286, 1302 (11th Cir. 2001), called into doubt on other grounds by Regalado Cuellar v. United States, 553 U.S. 550, 556 n.1, 128 S. Ct. 1994, 1999 n.1 (2008). The appellants have identified no grounds to support overturning the District Court’s conclusion that none of the jurors had in fact sent that letter to the Court. Finally, with respect to the first letter the jurors received, the appellants also argue that the District Court should have further investigated one of the juror’s allegation that she saw a fellow juror’s name on her letter. However, the District Court did question the juror whose name purportedly appeared on the letter, and that juror said that she did not send that letter. We cannot, on this record, overturn the District Court’s implicit determination that the juror was credible in her response. See Abbell, 271 F.3d at 1302. Neither can we say that the District Court abused its discretion in choosing not to further investigate the matter. See Yonn, 702 F.2d at 1345. J. REMAINING EVIDENTIARY RULINGS Tobin argues that the government failed to provide notice of its intent to cross-examine him on an alleged disciplinary action taken by the South Carolina Medical Board in March 2004. Tobin contends that this violation of Rule 404(b) 71 requires the reversal of his convictions. Tobin concedes that plain error review applies because he did not raise an objection when the government asked the question at trial. We conclude that any error did not affect his “substantial rights.” United States v. Lejarde-Rada, 319 F.3d 1288, 1289 (11th Cir. 2003). As noted, Tobin was asked about disciplinary actions taken by the medical boards of Texas and North Carolina. The brief questioning about whether he had also been disciplined by the South Carolina Medical Board did not add much to the testimony that the jury had already heard. LaCour and Tobin also challenge other of the District Court’s evidentiary rulings. We have carefully reviewed the record and “find all of [these] contentions to be meritless and without sufficient weight to require discussion.” Frink, 912 F.2d at 1415.22 K. PROSECUTORIAL MISCONDUCT Tobin and Baranwal argue that the government engaged in misconduct during closing argument and that the District Court’s inadequate response requires the reversal of their convictions. A conviction may be reversed on the basis of prosecutorial misconduct only if the prosecutor’s remarks 1) were improper and 2) 22 This observation applies equally to LaCour’s separate arguments that the District Court erred in denying his request for access to the transcripts of the grand jury proceedings and his request for Jencks Act material for one of the government’s witnesses. 72 prejudiced the defendant’s substantial rights in light of any curative instruction. United States v. O’Keefe, 461 F.3d 1338, 1350 (11th Cir. 2006). “A defendant’s substantial rights are prejudicially affected when a reasonable probability arises that, but for the prosecutor’s statements, the outcome of the trial would have been different.” Id. “When the record contains sufficient independent evidence of guilt, any error is harmless.” United States v. Eckhardt, 466 F.3d 938, 947 (11th Cir. 2006). 1. Tobin During rebuttal argument, the government stated that the controlled substances at issue for Count 3 and Count 11 were dispensed by a company called Excelsior Pharmaceuticals. The government argued that employees of Excelsior were salaried and that as a result, they did not have tracking numbers for the orders. Tobin objected to this statement on the ground that the government was arguing facts not in evidence. The District Court overruled the objection, instructing the jury: “it’s up to you to recall the evidence [and to] rely upon your memory of the evidence. [If] counsel tells you something that you don’t recall, you should rely upon your memory, not counsel’s memory.” Tobin asserts that the government’s statement during closing argument requires his convictions to be reversed. According to Tobin, the government was 73 attempting to explain the absence of tracking numbers in order to persuade the jury that the controlled substances were in fact shipped. Tobin’s argument is unavailing because it rests on the assumption that the government was required to prove the actual shipment of the controlled substances in order to obtain a conviction under Section 841(a)(1). As we have explained, the CSA required the government only to show that Tobin made an “attempted transfer of a controlled substance.” 21 U.S.C. § 802(8) (emphasis added). Thus, even if the government made an improper argument, Tobin has not shown that it affected his substantial rights. 2. Baranwal During its case in chief, the government called Timothy Heider, a former reporter of the Cleveland Plain Dealer newspaper, as a witness. On direct examination, Heider testified that in August 2004, he was working on a story on Internet pharmacies when he placed an order on the Internet for a drug called Didrex. He testified that the website on which he placed the order asked several questions, but that he did not provide a prescription for that drug. Heider said he received a bottle of the drug with Baranwal’s name on the label. He also testified that he later contacted Baranwal, telling “him that I received a prescription drug, his name was on the label, and I asked him why.” Heider said that Baranwal 74 “denied that he issued the drugs and didn’t know how his name ended up on the label, how he got the drugs.” On cross-examination by Baranwal’s counsel, Heider testified that he wrote two stories on the subject of Internet pharmacies. He stated that “[o]ne story was about an Internet drug operation that had started in Ohio and migrated to Florida,” but that story involved websites that were different than “the one dealing with Dr. Baranwal.” He did not describe the other story. On cross-examination by LaCour, Heider again identified his “first story” as involving the “Internet operation that started in Ohio and [then] migrated to Florida.” Again, at that stage of his testimony, he did not elaborate on any other stories that he wrote or published. During the government’s rebuttal argument, it made the statement: “The reason why Akhil Baranwal left [Jive Network] on November 1, 2004, is because he had been outed by an investigative reporter who told you that he wrote a big article about this, and Dr. Baranwal was front and center in that article because he, this investigative reporter, got a prescription that had Akhil Baranwal’s name on it.” The government’s argument drew an objection from Baranwal. The District Court stated: “Well, the reporter testified about it. The jury can recall his testimony as an investigative reporter and the work that he did on that context, and take that into consideration.” 75 After the jury retired to deliberate, Baranwal asked the District Court to instruct the jury to disregard the government’s argument about the article. The Court took the motion under consideration. The following morning, the District Court spoke to the parties about this issue. The District Court expressed concern that the government’s argument did not reflect Heider’s testimony. Baranwal requested that the District Court instruct the jury to “disregard the argument of government counsel.” In response, the government argued that the District Court had previously instructed the jury not to consider argument by counsel as evidence. The District Court denied Baranwal’s requested instruction. We agree with Baranwal that the government argued facts that were not in evidence when it stated that Heider wrote a “big article” in which Baranwal was “front and center” and that this “outed” him. However, Baranwal has not carried his burden of showing that this prejudiced him. According to Baranwal, the government’s argument was prejudicial to him because it “goes to the heart of [his] defense that he believed in good faith his actions . . . were in keeping with professional practice.” This argument again ignores the fact that the standard by which the jury was required to determine whether Baranwal was acting in the “usual course of professional practice” is an objective one. With respect to the substantive counts—the only counts of which he was convicted—his subjective 76 good faith did not matter. In light of this, and given that Baranwal has not otherwise explained how the government’s argument prejudiced him, we reject Baranwal’s suggestion that his convictions be reversed. L. THE ALLEN CHARGE LaCour, Tobin, and Pickens also argue that their convictions must be reversed because the District Court’s Allen charge and its accompanying statements coerced the jury into rendering a verdict. The jury began deliberating around 3 p.m. on April 27, 2009. Around 2 p.m. on April 29, 2009, the District Court received a message from the jury, indicating that it could not “reach a unanimous agreement for each defendant on any and all counts.” In response to the jury’s note, the District Court gave an Allen charge, and the jury retired to continue its deliberations. Shortly before lunch on the next day, Thursday, April 30, 2009, the District Court observed that it had previously informed the jury that the Court would not be in session on Fridays or the week of May 4, 2009. The District Court asked the jurors to provide input as to whether they would like to deliberate either on Friday, May 1, 2009 or the following week. Following lunch, the jury returned a note to the District Court, stating that it had agreed not to meet on Friday and that a majority of jurors would not be able to meet the following week. A few hours 77 later—about twenty-four hours after the District Court gave the Allen charge—the jury informed the Court that it had reached a verdict. The appellants argue that the District Court erred by giving the Allen charge and that the District Court’s other comments had the effect of coercing the jury into rendering a verdict. Neither suggestion carries the day. The appellants assert that the short turnaround between the Allen charge and the verdict indicates that some jurors were pressured into giving up their honestly held beliefs. However, this Court has upheld the use of the Allen charge in cases where the turnaround time was significantly shorter. See United States v. Dickerson, 248 F.3d 1036, 1050 (11th Cir. 2001) (citing Andrews v. United States, 309 F.2d 127, 129 (5th Cir. 1962) (verdict returned twenty-five minutes after the charge), and United States v. Chigbo, 38 F.3d 543, 545–46 (11th Cir. 1994) (verdict returned fifteen minutes after the charge))). Neither is the appellants’ argument that the District Court’s remaining comments had the effect of coercing the jury persuasive. Contrary to the appellants’ statement, the District Court did not tell the jury that it was going to keep them for another week. Rather, the District Court reminded the jury that it was not planning to be in session the week of May 4, 2009. The District Court then gave the jury the option of deliberating that week, instead of coming in the 78 week of May 11, 2009. We cannot say that the District Court coerced the jury into returning a verdict by giving them the option of continuing deliberations a week sooner than it had originally planned. M. THE JURY’S QUESTION During the lunch break on April 30, 2009, the jury sent a note to the District Court, asking “Your Honor—Can we the jury please see you privately as a group?” The District Court responded at 12:45 p.m., stating “Please outline the subject matter you wish to discuss and also assure me that every member of the jury is in agreement with the request.” The jury then wrote Your Honor—We the Jury are having difficulties coming to an agreement on the “U.S. generally recognized and accepted usual course of practice” guidelines for physicians and pharmacists in the internet medicine industry—1st paragraph on page 41. A unanimous decision has not been concluded. The District Court then “spent some time trying to figure out how [it] was going to answer that question and while [it] was still attempting to do that,” the jury sent another note, stating “Your Honor—We are okay to continue with our deliberations.” The District Court then asked “Do you or don’t you want an answer to this question?” The jury responded, “We do not need an answer. We are okay. Thank you.” Around 1:50 p.m., the parties returned to the courtroom, 79 and the District Court disclosed the jury communications to them, indicating that it “did not answer that question about the standard.” LaCour, Tobin, Baranwal, and Pickens contend that the District Court committed reversible error by not immediately disclosing the jury’s question to the parties. This argument is without merit. We have recognized that under Rule 43 of the Federal Rules of Criminal Procedure, “a message from a jury should be answered in open court after counsel has been informed of its substance and has been given an opportunity to be heard.” United States v. Rapp, 871 F.2d 957, 966 (11th Cir. 1989), abrogated on other grounds by United States v. Wells, 519 U.S. 482, 117 S. Ct. 921 (1997). We have also held, however, that a district court’s failure to do so is not reversible error if it is harmless. Id. In this case, any failure on the part of the District Court to disclose the jury’s message was harmless because the jury ultimately indicated that it no longer needed an answer to its question. The appellants suggest that the delay by the District Court in responding to the question may have prompted the jury to give up on getting an answer; they imply that the District Court delayed responding in order to avoid answering the question. It is true that “[a] trial judge has some obligation to make reasonable efforts to answer a question from the jury.” United States v. Rodriguez, 765 F.2d 80 1546, 1553 (11th Cir. 1985). On this record, however, we cannot say that the District Court failed to undertake reasonable efforts. Beyond that, we note that when the jury told the District Court that it did not need an answer to its question, the Court double-checked with the jury to see if it wanted a response, and the jury indicated that it did not. Under these circumstances, the appellants have not shown reversible error. Cf. id. at 1554 (finding no reversible error where “district court was not able to respond to the jury’s question before the jury rendered a verdict”). N. CUMULATIVE ERROR LaCour, Tobin, Baranwal, and Pickens argue that the cumulative errors committed by the District Court deprived them of a fair trial. We “may consider the cumulative effects of errors to determine if [a] defendant has been denied a fair trial.” United States v. Ladson, 643 F.3d 1335, 1342 (11th Cir. 2011) (quotation marks omitted). “There is no cumulative error where the defendant cannot establish that the combined errors affected his substantial rights.” Id. (quotation marks omitted). We have carefully reviewed the record and conclude that although the District Court made errors in this case, they are few, and taken together, they have not affected the substantial rights of the appellants. O. JUDICIAL PARTICIPATION IN PLEA DISCUSSIONS 81 Rule 11(c)(1) of the Federal Rules of Criminal Procedure provides that “[t]he court must not participate in [plea] discussions.” Fed. R. Crim. P. 11(c)(1).23 In United States v. Adams, 634 F.2d 830 (5th Cir. 1981), the former Fifth Circuit held that this rule imposes “an absolute prohibition on all forms of judicial participation in . . . the plea negotiation process.” Id. at 835. The Court noted that this rule serves important purposes. Specifically, where the defendant refuses to plead guilty and goes on to trial, the rule helps to avoid “two serious risks in the subsequent trial and sentencing.” Id. at 840. First, because a trial judge’s actual impartiality may be impaired as a result of having participated in plea discussions, the rule serves to safeguard the trial judge’s actual neutrality. Id. Second, because participation may give the impression that the trial judge is no longer a neutral arbiter, the rule also serves to protect the appearance of impartiality. Id. at 841. Because these concerns are of a fundamental nature, the Court in Adams held that judicial participation in plea discussions constitutes plain error that this Court can raise sua sponte pursuant to its supervisory power over the district 23 Prior to being amended in 2002, Rule 11 stated that “the court shall not participate in any [plea] discussions.” See, e.g., United States v. Corbitt, 996 F.2d 1132, 1134 (11th Cir. 1993) (citing the previous language). The change in language is stylistic and does not alter the substantive law. See Fed. R. Crim. P. 11 advisory committee’s note. 82 courts. Id. at 836. The Adams Court also held that it is not necessary for a defendant to show that he was actually prejudiced by such conduct. Id. Finally, the Court held that as a remedy for a violation of the rule, a defendant who has not pleaded guilty and who does not show actual prejudice at trial or sentencing “should not receive a new trial but should be resentenced before a different judge.” Id. at 842.24 In United States v. Corbitt, 996 F.2d 1132 (11th Cir. 1993), the district court stated during a status conference that “[defendants] want to go out and get arrested, they come in here and they’ll get a fair trial, and if they get found guilty, they’ll also get a fair sentence, fairly high.” Id. at 1133–34. We indicated that the district court’s statement was not “flagrant,” but nonetheless concluded that the district court had violated Rule 11. Id. at 1134–35. We reiterated the holdings of Adams: that Rule 11 establishes “an absolute prohibition on all forms of judicial participation,” that such participation is “plain error that can be raised on appeal sua sponte,” and that the defendant need not show actual prejudice to obtain a remedy. Id. at 1134 (quotation marks omitted). 24 Prior to Adams, the Supreme Court in McCarthy v. United States, 394 U.S. 459, 89 S. Ct. 1166 (1969), addressed the question of the appropriate remedy in a case where the defendant does plead guilty following the district court’s participation in plea discussions. See id. at 464, 89 S. Ct. at 1169. The Supreme Court held that such a defendant should be allowed to withdraw his guilty plea. Id. 83 In United States v. Casallas, 59 F.3d 1173 (11th Cir. 1995), the defendant informed the district court at the beginning of a change-of-plea hearing that he no longer wished to plead guilty. Id. at 1176. After some discussion about the charges that the defendant faced, the district court said: “I suggest to the Defendant that he talk to his lawyer some and see if [not pleading guilty] is really what he wants to do.” Id. We observed that “these comments were innocuous and intended only to insure that [the defendant] was making an informed decision.” Id. at 1177. We nonetheless concluded that the district court “crossed the line.” Id. at 1178. Adhering to the “bright-line rule” that “prohibits the participation of the judge in plea negotiations under any circumstances,” we made it clear that we will not “become involved in evaluating the degree of judicial participation.” Id. at 1177, 1178 (quotation marks omitted).25 In United States v. Diaz, 138 F.3d 1359 (11th Cir. 1998), the district court told the defendant right before trial: “I’m glad to go to trial here, I’ve got the jurors outside, we’re going to trial. There’s no problem about that.” Id. at 1361. 25 In United States v. Johnson, 89 F.3d 778 (11th Cir. 1996), we declined to conclude that there was a Rule 11 violation. Id. at 782–84. Our decision in that case rested on the unusual fact that the statements that allegedly violated Rule 11 were made by the district court during the middle of a change-of-plea colloquy and that the defendant did not “accept the court’s invitations to break off the colloquy.” Id. at 784. We noted that the district court’s statement that the defendant speak to his attorney was in response to the defendant’s “vacillations” about the change of plea and that it was an attempt on the part of the district court to determine whether the colloquy should end. Id. at 783. 84 The district court also added, however, “But you need to think about [yourself] . . . and you’re going to risk ten years in prison, you need to think about your options. You know, I’ll be glad to sit here, we’re glad to try your case, but . . . the evidence is kind of compelling.” Id. at 1361–62. The defendant refused to plead guilty and was convicted of all counts. Id. at 1362. On appeal, the defendant requested a new trial or, in the alternative, re-sentencing before a different judge. Id. at 1363. We held that the district court’s comments violated Rule 11, but that the defendant was not entitled to a remedy. Id. We acknowledged that Adams authorizes re- sentencing before a different district judge even in the absence of prejudice. Id. at 1364. However, we found Adams distinguishable because at the time that Adams was decided, there were no sentencing guidelines, and the district judge thus had significant discretion in fashioning a sentence. See id. We noted that in Diaz’s case, the sentencing guidelines constrained the district judge. Id. In this case, the District Court addressed the issue of plea negotiations on at least two occasions. First, during a status conference on February 4, 2009, the District Court stated that “sentencing consequences are always very important” and that as a result, it “requested that the government be prepared to enlighten [defense counsel] as to where counsel believed that [the] individual defendant was situated with respect to the guidelines.” While the District Court stated that it did 85 not intend to “engage in plea negotiations,” it said that it “would certainly like to see at least the process begin.” The District Court added: “sooner or later the doctors have to come to grips with what they’re looking at” in terms of possibly losing their licenses. The District Court stated that “in my view, what accentuates the need to have some consideration of the impact of the sentencing guidelines is the possibility that one or more of the doctors may lose his or her license upon a conviction.” The District Court then asked the government to identify the status of any plea discussions. During that colloquy, the District Court noted again that “if I were defense counsel I would like to know” the applicable sentencing guidelines range, “assuming the plea of guilty with acceptance is one thing, a conviction with no acceptance is another thing.” After addressing the government, the District Court then said that it would like to “hear from [defense counsel] one by one” as to whether they had relayed the government’s information to their clients and otherwise “looked at the question of what the impact is on the continued license to practice.” Counsel for Tobin, Pickens, Baranwal, Chebssi, and other defendants then informed the District Court of the status of the plea discussions and some of the basic considerations underlying those discussions. LaCour was represented by counsel at the time, but he did not discuss the issue with the District Court. 86 In the middle of that discussion, albeit with respect to a defendant who eventually pleaded guilty, the District Court asked the government: “Does the guideline calculation range change for the betterment of the defendant if not convicted of the conspiracy, but convicted of one or more of the individual counts?” The District Court then turned to defense counsel and said, “I guess the thing I would be curious if I were in your shoes representing your client is whether the people who have her license at issue would treat her more lightly if she was convicted of individual counts and not the conspiracy.” Finally, in the middle of the discussion, the District Court also noted: “I might clear up a matter that [the previous District Judge to whom the case was assigned] and I have talked about this, and I think it’s our collective view that I should be the one that sentences everybody.” The second time the District Court addressed the issue of plea negotiations was during a status conference on February 25, 2009. The District Court stated, “I want to now address each of the defendants individually and talk about the possibility of other resolutions of this matter other than a trial.” The District Court acknowledged that it is “strictly prohibited from attempting to work out any kind of an agreement upon resolution of the case short of a trial.” The Court nonetheless went on to note that “[t]his case if there is a conviction requires the 87 Court [to] determine where that defendant fits with respect to the sentencing guidelines.” The Court then distributed a chart that, based on the record, may have shown the sentencing guidelines ranges, and it went on to discuss the history of the federal sentencing regime and the applicable law as it stands. In describing the current sentencing regime, the District Court stated that “the Supreme Court has made it pretty clear . . . that we judges have a great deal of discretion in that area.” The District Court did state, “Remember, I’m not permitted to engage in plea negotiations, so I’m not suggesting to any defendant that you should consider pleading guilty.” The District Court, however, said: “But I am concerned that you understand what the position of the government is if in fact you should want to think about pleading guilty. . . . I’m not interested in the details [of any offer], but I want to be sure that each defendant for whom the government has presented a proposition knows about the proposal.” The District Court then asked the government to indicate whether it had presented a proposal to the defendants. The government stated that it did put forward a proposal for each defendant. The District Court then directly asked each individual defendant, including all five appellants, whether they had seen the proposal. 88 It is clear that in light of all of these comments, the District Court violated the “categorical mandate” of Rule 11, as LaCour suggests. Corbitt, 996 F.2d at 1134. The government resists this conclusion by noting that none of the appellants made any objections at the time of these discussions. The government’s attempt to argue that this affects the scope and standard of our review is clearly misplaced. Again, we have recognized that judicial involvement in plea discussions is an issue that implicates the integrity of criminal proceedings and therefore, it constitutes plain error we may consider sua sponte. See Adams, 634 F.2d at 836; see also Corbitt, 996 F.2d at 1134. The government acknowledges that the District Court’s comments were “inadvisable,” but in so doing, seems to suggest that they are consistent with the requirement of Rule 11. We cannot agree. Rule 11 imposes “an absolute prohibition on all forms of judicial participation in . . . the plea negotiation process.” Adams, 634 F.2d at 835. This is a bright-line rule, and we have made it clear that we will not engage in the exercise of determining the degree to which a district court discusses the subject of plea negotiations. Casallas, 59 F.3d at 1178. Here, the District Court’s comments “crossed the line.” Id. The District Court explicitly indicated that it would like the defendants to begin and engage in plea discussions. The District Court addressed both defense counsel and the 89 defendants themselves in individual colloquies on this subject. During that time, the District Court reminded the parties that it would be sentencing the defendants and that it has a “great deal of discretion” in that regard. The District Court also went to great lengths to make sure that the defendants understood the sentencing guidelines for the counts with which they were charged. The District Court also repeatedly drew attention to additional repercussions that could flow from a conviction on all counts, such as the potential loss of the defendants’ professional licenses. The government emphasizes that the District Court itself believed that it was acting in a way that is consistent with Rule 11 and that the District Court did not comment on the strength of the government’s case or discuss specific sentences. We likewise observe that the record indicates that the District Court was well-intentioned in engaging in this discussion. The Court was evidently trying to ensure that the defendants had all of the relevant information so that they could make an informed decision about how to proceed with the case. However, we have made it clear that Rule 11 can be violated even when a district court’s comments are “innocuous and intended only to insure that [the defendant is] making an informed decision.” Casallas, 59 F.3d at 1177. Indeed, “while a court may be motivated primarily by the concern that [the defendant] be thoroughly 90 apprised of the situation that he face[s], . . . this concern, however well- intentioned, [does] not excuse judicial participation.” United States v. Davila, 664 F.3d 1355, 1359 (11th Cir. 2011) (quotation marks omitted). The fact that the District Court did not comment on the government’s case or on specific sentences is also immaterial. Again, the rule imposes “an absolute prohibition on all forms of judicial participation in . . . the plea negotiation process.” Adams, 634 F.2d at 835. We have refused to determine the degree to which a district court has discussed the subject of plea negotiations. Casallas, 59 F.3d at 1178. Simply put, district courts should not offer any comments “touching upon” this subject. Diaz, 138 F.3d at 1363. Here, the District Court’s extensive comments, including its statement that it would like to see the process of plea negotiations begin, raise a key concern that underlies Rule 11—namely, that a district court’s suggestion that a defendant look into pleading guilty may give the impression that the district court has already taken a position regarding the question of guilt and that this can undermine the defendant’s confidence in the neutrality of the tribunal and the fairness of the subsequent proceedings. Adams, 634 F.2d at 841. The government argues that even if the District Court impermissibly discussed the subject of plea negotiations, LaCour is not entitled to any remedy. 91 The government’s suggestion is correct to some extent. The idea that LaCour is entitled to a new trial is clearly foreclosed by precedent. The Court in Adams held that a defendant who has not pleaded guilty and who does not show actual prejudice at trial or sentencing “should not receive a new trial.” See Adams, 634 F.2d at 840. LaCour has not carried his burden of showing that he was actually prejudiced at trial or sentencing. Thus, we reject his request that we reverse his convictions and remand the case for a new trial based on the Rule 11 violations. However, the government’s argument fails to the extent that it suggests LaCour is not entitled to a re-sentencing before a different district judge. The government rests its argument on our decision in Diaz. There, we held that the remedy of re-sentencing, which was announced in Adams, was not necessary. See 138 F.3d at 1363. We found that Adams was distinguishable because at the time it was decided, there were no sentencing guidelines, and the district court had great discretion in sentencing defendants. Diaz, 138 F.3d at 1364. In Diaz, we found it comforting that by contrast, the district court was constrained by the sentencing guidelines. Id. The government’s argument falls flat because following Diaz, the Supreme Court decided United States v. Booker, 543 U.S. 220, 125 S. Ct. 738 (2005). There, the Court rendered the sentencing guidelines advisory, see id. at 245, 125 92 S. Ct. at 756–57, and as a consequence, district courts now have significant discretion in sentencing criminal defendants. Indeed, a district court can impose a sentence outside of the guidelines range, and its decision to do so is reviewed only for abuse of discretion. See Gall v. United States, 552 U.S. 38, 51, 128 S. Ct. 586, 597 (2007). As the Supreme Court has observed, “the range of choice” has been “significantly broadened.” Id. at 59, 128 S. Ct. at 602 (quotation marks omitted). The Supreme Court’s decision in Booker has thus transformed the landscape, and it is apparent to us that as a result, the very reasoning of our decision in Diaz requires us to return to the rule announced in Adams.26 Indeed, given that significant discretion has been restored to the district courts in matters of sentencing, we are no longer confident as we were when we decided Diaz that a district court’s participation in plea discussions will not affect the court’s actual impartiality or the appearance of impartiality at sentencing, both of which are fundamental to the integrity of that process. See Adams, 634 F.2d at 840–41. Thus, adhering to the rule announced in Adams, we conclude that LaCour’s 26 The government also draws attention to the Diaz Court’s observations that in that case, the district court “merely engaged in a straightforward discussion of the applicable guidelines” and the sentence was based on “entirely sound reasons,” evincing “no bias” toward the defendant. Diaz, 138 F.3d at 1364. Under Adams, however, the presence or absence of actual bias or prejudice is irrelevant. See 634 F.2d at 842; see also Davila, 664 F.3d at 1358 (“Notably, while other circuits recognize harmless error in the context of judicial participation, we do not.”). Under Adams, even when a defendant cannot show actual bias or prejudice, he is entitled to the remedy of re-sentencing before a different district judge. 634 F.2d at 842. 93 sentence must be vacated and that a different district judge must be assigned to conduct his re-sentencing.27 P. REMAINING SENTENCING ISSUES Baranwal and Chebssi also challenge their sentences, but none of the arguments they raise have merit. 1. Use of Acquitted Conduct in Sentencing Baranwal and Chebssi were both acquitted of Count 1, which charged them with conspiracy under 21 U.S.C. § 846. On appeal, they argue that the District Court erred in taking into account the conduct for which they were acquitted in fashioning their sentences. As the government points out, however, our precedent authorizes a sentencing judge to “consider relevant acquitted conduct” that has been proven by a preponderance of the evidence, United States v. Duncan, 400 F.3d 1297, 1305 (11th Cir. 2005), and we have held that this does not violate a defendant’s right to trial by jury or a defendant’s due process right to have facts be proven beyond a reasonable doubt, see, e.g., United States v. Belfast, 611 F.3d 27 LaCour has made two other arguments regarding the propriety of his sentence—that the District Court erred in granting in part the government’s motion for upward departure and in denying his request for a two-level offense level reduction for acceptance of responsibility. In light of our decision to vacate his sentence, we do not address these additional issues. Tobin and Pickens appear to have adopted LaCour’s Rule 11 argument, but on appeal, they have not challenged their sentences in any way. We therefore find it appropriate not to vacate their sentences. 94 783, 827 (11th Cir. 2010); United States v. Faust, 456 F.3d 1342, 1347 (11th Cir. 2006); United States v. Barakat, 130 F.3d 1448, 1452 (11th Cir. 1997). Chebssi argues that allowing a judge to consider acquitted conduct deprives her of the due process right to fair notice. We have not squarely addressed this argument before, but find it to be without merit. In United States v. Irizarry, 458 F.3d 1208 (11th Cir. 2006), we observed that “[a]fter Booker, parties are inherently on notice that the sentencing guidelines range is advisory.” Id. at 1212. Similarly, given that “our precedents have consistently upheld district courts’ use of [a defendant’s] relevant conduct when determining an appropriate sentence,” United States v. Campbell, 491 F.3d 1306, 1314 (11th Cir. 2007), it is apparent to us that parties in this Circuit, such as Chebssi, are inherently on notice that a sentencing judge may take into account such conduct as well. 2. Procedural Reasonableness A district court should ensure that it commits no procedural error in imposing sentence. United States v. Saac, 632 F.3d 1203, 1212 (11th Cir. 2011). Procedural errors include “selecting a sentence based on clearly erroneous facts” and “treating the [Sentencing] Guidelines as mandatory.” Id. Baranwal argues that the District Court failed to make a specific finding regarding the relevant conduct that it took into account in fashioning his sentence. 95 “We recognize that, to facilitate appellate review, a district court should make explicit factual findings that underpin its sentencing decision.” Bradley, 644 F.3d at 1293. But the failure to make specific findings does not preclude meaningful appellate review if the court’s sentencing decision is based on “clearly identifiable evidence.” Id. In this vein, Baranwal argues that the record does not contain such evidence. However, his one-sentence argument does not cite to any portions of the record or otherwise provide any elaboration. We therefore find this argument to be waived. See Carmichael v. Kellogg, Brown & Root Servs., Inc., 572 F.3d 1271, 1283 (11th Cir. 2009). Chebssi asserts that her sentence is procedurally unreasonable because the District Court believed that the sentencing guidelines are mandatory. In support of this argument, Chebssi cites to the District Court’s statement that “I’m required to pronounce sentences that I find to be required by the law.” This argument misunderstands the District Court’s observation. In light of the Court’s discussion of the subject of guilty pleas, it is evident to us that the Court was manifestly aware of the advisory nature of the sentencing guidelines. The District Court’s statement can only be understood as underscoring its view that it was obligated to impose a sentence in accordance with the factors outlined in 18 U.S.C. § 3553(a). 3. Substantive Reasonableness 96 Chebssi contends that her sentence is substantively unreasonable, and Baranwal also appears to make this argument. Here, the District Court found that the guidelines range for Chebssi was 33 to 41 months. The District Court granted a downward variance, lowering the range to 15 to 21 months, and imposed a sentence of 15 months imprisonment. The District Court found that the guidelines range for Baranwal was 41 to 51 months. The District Court also granted a downward variance, lowering the range to 27 to 33 months and sentenced Baranwal to 27 months imprisonment. We have observed that “[a]lthough we do not automatically presume a sentence within the guidelines range is reasonable, we ‘ordinarily . . . expect [such] a sentence . . . to be reasonable.’” United States v. Hunt, 526 F.3d 739, 746 (11th Cir. 2008) (quoting United States v. Talley, 431 F.3d 784, 788 (11th Cir. 2005)). In light of this recognition, and in light of the significant downward variances granted by the District Court to both Chebssi and Baranwal, we are simply unable to say that the sentences are substantively unreasonable. Chebssi and Baranwal draw our attention to the fact that other defendants in the case received less serious sentences. They have not shown, however, that the District Court thereby created “unwarranted sentence disparities.” 18 U.S.C. § 3553(a)(6). Indeed, as Baranwal himself acknowledges, these other defendants 97 pleaded guilty. The fact that these defendants received lower sentences does not, in and of itself, indicate that any sentencing disparities were unwarranted. See United States v. Mateos, 623 F.3d 1350, 1367 (11th Cir. 2010) (noting that “there is no unwarranted disparity when a cooperating defendant pleads guilty and receives a lesser sentence than a defendant who proceeds to trial” (quotation marks omitted)). IV. CONCLUSION For the foregoing reasons, we affirm the convictions of all five appellants, as well as the sentences of Baranwal and Chebssi. We vacate LaCour’s sentence and remand for re-sentencing before a different district judge. AFFIRMED IN PART, VACATED IN PART, AND REMANDED. 98