United States v. Shah

Before Richman, Chief Judge, and Wiener and Willett, Circuit Judges. Priscilla Richman, Chief Judge: No member of the panel nor judge in regular active service of the court having requested that the court be polled on rehearing en banc (Fed. R. App. P. 35 and 5th Cir. R. 35), the petitions for rehearing en banc are  DENIED. But, the panel’s opinion issued October 2, 2023,[1] is WITHDRAWN, and the following opinion is SUBSTITUTED. Seven codefendants appeal their various convictions stemming from a multi-million-dollar healthcare conspiracy involving surgery-referral kickbacks at Forest Park Medical Center in Dallas, Texas. They challenge convictions under the Anti-Kickback Statute (which will sometimes be referred to as AKS),[2] the Travel Act,[3] and for money laundering.[4] Finding no reversible error, we affirm the district court’s judgment. I The seven codefendants on appeal were all convicted of engaging in a $40 million healthcare conspiracy in Dallas, Texas. Our initial discussion of the facts is limited to the general outline of the conspiracy: its origins, its major players, and its operation. We reserve a more detailed discussion of the evidence against the defendants for the sections of this opinion that deal with those facts more directly.[5] There are three main sets of actors in this case: the staff at Forest Park Medical Center (Forest Park or the hospital), surgeons Forest Park paid to perform surgeries at its hospital, and pass-through entities affiliated with both Forest Park and the surgeons. The defendants in this case are, with three exceptions, the surgeons whom Forest Park paid to direct surgeries to the hospital—Won, Rimlawi, Shah, and Henry. One exception is Forrest—she is a nurse. Another is Jacob—he ran Adelaide Business Solutions (Adelaide), a pass-through entity. The other is Burt—he was part of the hospital’s staff. But this case begins with three men who are not parties to the current appeal—Alan Beauchamp, Wade Barker, and Richard Toussaint. They decided to open a hospital together—Forest Park. Forest Park was to be an “out-of-network” hospital, meaning that it was not affiliated with any insurance carrier and any surgeries performed there would be considered out-of-network for the patients. They planned for their hospital to be out-of-network because insurers were reimbursing out-of-network facilities at very high rates. But they faced a difficulty: how to convince patients to pay out-of-network costs when they could have the surgery performed at an in-network facility? Their answer: pay surgeons to refer patients to Forest Park and then waive the patient’s financial responsibility beyond what the surgery would cost in-network. In creating such a structure, the Government asserts that Forest Park engaged in illegal conduct. First, the hospital was “buying surgeries,” i.e., it paid surgeons to perform a surgery at the hospital. It is well established that buying surgeries is illegal, as many witnesses testified.[6] Second, the hospital’s formal internal policy was not to waive patient financial responsibility. So, the Government argues, Forest Park’s upper management had to cover its tracks. It did this by creating or partnering with a number of pass-through entities to create sham marketing or consulting contracts with the surgeons. One such entity was Adelaide, overseen by defendant Jacob. Another was Unique, which was operated by Beauchamp, Andrea Smith (a longtime aid to Beauchamp), and defendant Burt. The Government argued that the conspiracy was as follows: The hospital and surgeons reached an agreement whereby the hospital would pay the surgeons to refer patients to Forest Park; the hospital would then contract with a pass-through entity for sham marketing or consulting services; the surgeons would contract with the same pass-through entity for sham marketing or consulting services as well; the surgeons would then direct their patients to Forest Park for surgery; Forest Park would obtain reimbursements from insurers at the out-of-network rate; the hospital would pay the pass-through entities some of those profits; and then the pass-through entities would pass along those profits to the surgeons for marketing and consulting services the surgeons never rendered. Although Forest Park employed legitimate hospital staff, it also employed a number of individuals in roles relating directly to the conspiracy. Andrea Smith’s role was to keep track of all the surgeries that the hospital “bought” and make sure that the surgeons were reimbursed according to the rates they had agreed to. She created detailed spreadsheets to keep track of this, and those spreadsheets became a major part of the Government’s case. Burt’s job was to assist Beauchamp in recruiting surgeons and patients. Along with Beauchamp and Smith, Burt formed an organization called Unique that was a pass-through entity. Eventually the controller for Forest Park began to resist doing business with Unique. The hospital’s leadership team decided to create an outside group. Jacob owned a radiology company near the hospital. He and Beauchamp were friends. Beauchamp approached Jacob to join the enterprise, and Jacob agreed. Jacob formed Adelaide, which assumed the role of the pass-through entity formerly occupied by Unique. Forest Park paid Adelaide monthly for services that Adelaide never rendered to the hospital. Instead, Beauchamp sent a monthly check to Adelaide with specific instructions as to how Jacob was to pay the surgeons he “contracted” with for marketing or consulting services. Often, the surgeons would complain they had not been reimbursed at their agreed-upon rate. Won, Rimlawi, Shah, and Henry are surgeons who contracted with a pass-through entity for marketing or consulting services and who directed some of their patients to Forest Park. Most of these patients had private insurance, but some of them were covered by a federal healthcare program including Medicare, TRICARE, or DOL/FECA. Forest Park then paid the surgeons with checks issued through the pass-through entity. Forrest is a nurse who was involved in the scheme, who at the time persuaded patients to have their surgery performed at Forest Park. The district court’s description is apt: “[O]nce you separate all the ‘noise,’ the trial involved a single pyramid conspiracy with a number of participants. . . . Attempts were made to paper their dishonest conduct—to hide behind sham contracts—which ultimately proved unsuccessful.” The defendants who are parties to this appeal were tried together. The jury convicted all but Burt for engaging in a conspiracy that violated the Anti-Kickback Statute.[7] The jury convicted Jacob, Shah, Burt, Rimlawi, and Forrest of substantive violations of that statute. It convicted Henry and Burt on substantive violations of the Travel Act[8] as well as conspiring to commit money laundering. The jury acquitted a surgeon who is not a party to this appeal and failed to reach a verdict as to another. Shah was sentenced to 42 months of imprisonment; Rimlawi was sentenced to 90 months; Jacob to 96 months; Burt to 150 months; Henry to 90 months; Won to 60 months; and Forrest to 36 months. The defendants timely appealed. The defendants raise many of the same issues on appeal, often adopting each other’s arguments. We have organized this opinion into eighteen Parts following this one. This reflects the lowest combined count of the defendants’ various issues. In each part, we address the various arguments each defendant makes regarding a particular issue, including closely related sub-issues where appropriate. We begin with the defendants’ sufficiency of the evidence challenges. Then we address the remaining issues: whether the Texas Commercial Bribery Statute[9] is a proper predicate offense to a violation of the Travel Act; potential Speedy Trial Act[10] and Court Reporter Act[11] violations; purported violations of Burt’s proffer agreement and any Bruton[12] error stemming therefrom; various challenges to district court evidentiary rulings, jury instructions, and prosecutor arguments; and finally, challenges to sentencing and restitution. II Six defendants (all but Burt) challenge the sufficiency of the evidence supporting their respective convictions of conspiring to violate the AKS. The AKS provides, in relevant part, that: (1) [w]hoever knowingly and willfully solicits or receives any remuneration (including any kickback, bribe, or rebate) directly or indirectly, overtly or covertly, in cash or in kind . . . in return for referring an individual to a person for furnishing . . . of any item or service for which payment may be made in whole or in part under a Federal health care program, . . . shall be guilty of a felony . . . (2) [w]hoever knowingly and willfully offers or pays any [such] remuneration . . . to induce [such a referral] . . . shall be guilty of a felony.[13] These six defendants were convicted of engaging in a conspiracy to violate the AKS. To prove a conspiracy, the prosecutors had to show: (1) an agreement between two or more persons to pursue an unlawful objective; (2) that the defendant knew of the unlawful objective and voluntarily joined the conspiracy; and (3) an overt act done by one or more members of the conspiracy in furtherance of the conspiracy’s objective.[14] The degree of criminal intent necessary to sustain a conviction of conspiracy is the same as to sustain a conviction of the underlying offense.[15] To prove a violation of the AKS, the Government must prove that the defendant acted willfully, that is, “with the specific intent to do something the law forbids”[16] or “with bad purpose either to disobey or disregard the law.”[17] We review sufficiency of the evidence challenges de novo, but we remain “highly deferential to the verdict.”[18] “[T]he relevant question is whether, after viewing the evidence in the light most favorable to the prosecution, any rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt.”[19] “We will not second guess the jury in its choice of which witnesses to believe.”[20] A. Won Won argues that there was insufficient evidence to prove that he agreed to violate the AKS and that he willfully sent federal patients to Forest Park—arguing that the government had to prove that he knew the patients he sent were federally insured. The Government contends that Won misconstrues the AKS and that the Government did not need to prove that Won knew his patients were federally insured. 1 First, and as an apparent matter of first impression, this court must decide whether a conviction under the AKS requires the defendant to have knowledge that payment for the surgeries he referred “may be made in whole or in part under a Federal healthcare program.”[21] The Government argues that the “Federal healthcare reference” in the statute is simply the hook upon which jurisdiction is based and that, under well-settled precedent, it need not prove scienter as to the jurisdictional element. Jurisdictional elements “simply ensure that the Federal Government has the constitutional authority to regulate the defendant’s conduct.”[22] The Government is not required to prove mens rea for those elements.[23] Won argues that the federal healthcare program provision is not a jurisdictional hook, but a substantive element of the crime for which the Government had to prove intent. A Maryland district court has addressed this question and decided that the federal healthcare program requirement is a jurisdictional hook.[24] The Eleventh Circuit has addressed the question as well, and that court also appeared to consider the requirement of a federal healthcare program to be jurisdictional.[25] In Ruan v. United States,[26] the Supreme Court vacated the Eleventh Circuit’s decision on other grounds. The Supreme Court did, however, discuss the scienter requirement in a statute. The Court concluded that “knowingly” “modifies not only the words directly following it, but also those other statutory terms that ‘separate wrongful from innocent acts.’”[27] We note that as a general proposition, “buying” surgeries is not “innocent” conduct. That conduct is illegal under a number of states’ laws, and no party disputes that.[28] Nevertheless, in Ruan the Court said that “knowingly” also “modifies . . . the words directly following it.”[29] Here, “Federal healthcare programs” follows “knowingly.” At the very least, the federal healthcare reference in this statute clarifies to which “item[s] or service[s]” the statute applies. The question remains, does “ knowingly” apply to “item[s] or service[s].” We think that Won overlooks a key clause in the AKS. The AKS requires only that payment “may” be made by a federal healthcare program.[30] In United States v. Miles[31] we characterized that as meaning only that “an item or service . . . could be paid for by a federal health care program.”[32] Further support for this proposition is found in the AKS itself, which provides that “a person need not have actual knowledge of this section or specific intent to commit a violation of this section.”[33] So, contrary to Won’s argument, the Government did not have to show he knowingly referred federally insured patients for remuneration. All it had to show was that he knowingly agreed to accept remuneration for referring patients that could be federally insured. The Government met that burden. To the extent defendants argue they cannot be guilty because they intentionally avoided federally insured patients, they admit that they had agreed to accept remuneration for referring patients for services that could be paid for through a federal healthcare program. The Government did not need to prove Won knew he was referring federally insured patients. 2 The Government did need to prove that at least some patients were federally insured or that payment “may” have been made by a federal healthcare program—to establish federal jurisdiction.[34] The Government points to evidence that Won sent a TRICARE patient to Forest Park as well as tracking sheets showing Won received credit for Medicare patients. Won disputes both pieces of evidence. He argues that the TRICARE patient had TRICARE only as a backup and that Aetna actually paid for her surgery. He also argues that the tracking sheets showing Medicare patients were never referenced at trial. Even assuming that no TRICARE money changed hands, Won cannot nullify the Medicare evidence by claiming that it was never discussed at trial. The inquiry is whether a rational trier of fact could have found for the prosecution; we review the evidence, not the prosecution’s argument.[35] The evidence shows that Won referred some federally insured patients to Forest Park. Further, it shows that Won “want[ed] to discuss [with Beauchamp] the amount [his] surgeries [we]re going to be billed for and [the] expect[ed] . . . reimburse[ment].” The evidence also establishes that kickbacks were widely known to be illegal. A reasonable juror could have found an agreement between Won and Beauchamp to refer patients to Forest Park for remuneration, knowing that services to some of those patients might be paid, in whole or in part, under a federally funded healthcare program. This would satisfy the first two prongs of a conspiracy conviction.[36] Finally, the tracking sheets provide evidence that the referrals actually happened, satisfying the overt act element of a conspiracy conviction.[37] B. Rimlawi Rimlawi challenges the sufficiency of the evidence to support his conspiracy conviction for violations of the AKS on the grounds that there was no evidence that he received kickbacks for his four federal patients. Rimlawi argues that the evidence submitted to the jury established that the marketing agreements paid money only for “out-of-network” surgeries. He attempts to define “out-of-network” as excluding federal-pay surgeries. Under that theory, he argues, the jury could not infer that he received money for federally insured patients. At least on paper, the agreements sought to avoid federal-pay patients, but, regardless of what the paper agreement said, the question is whether the jury had enough evidence in front of it to infer that Rimlawi knowingly referred patients who may have been federal-pay patients. The Government argues that the tracking sheets, emails, and testimony of Beauchamp provide sufficient evidence to find that Rimlawi knowingly accepted payments “in return for referring an individual to a person for furnishing . . . of any item or service for which payment may be made in whole or in part under a Federal healthcare program.” Viewed in the light most favorable to the verdict, they do.[38] Beauchamp, for example, testified that Forest Park paid for federally insured patients. Rimlawi admits to having federally insured patients. Smith’s kickback tracking sheets show that Rimlawi was credited with DOL/FECA insured patients who are federal pay, and Rimlawi does not contest that DOL/FECA patients are federal pay. A jury could reasonably infer that Rimlawi received kickbacks for those patients and knew that payments might be made for at least some patients he referred by a federal healthcare program. C. Henry Henry essentially repeats Won’s and Rimlawi’s arguments. He claims that the jury did not have sufficient evidence to find that he accepted kickbacks for federal patients and that there was insufficient evidence to prove that he knew his DOL patients were federally insured. As to the former, Henry’s argument fails for the same reason as Rimlawi’s. There is evidence in the record that Henry sent DOL/FECA patients to Forest Park and received remuneration. Henry admits this. Henry’s second argument is stronger. He claims that in order for his conspiracy conviction to stand, the Government needed to prove that he knew his DOL patients were federally insured for purposes of the AKS. But this argument fails for the same reason that Won’s argument fails. The Government did not need to prove that the defendants knew their conduct targeted federal healthcare programs. It needed to prove that the defendants knew services to some patients they referred might be paid, in whole or in part, by a federal healthcare program. Additionally, as already noted, the AKS itself provides that “a person need not have actual knowledge of this section or specific intent to commit a violation of this section.”[39] Henry’s reliance on this court’s holding in United States v. Anderson[40] is misplaced. Henry cites that case for the proposition that to prove conspiracy to violate the AKS, the Government needed to prove that he entered the conspiracy with “the specific intent that the underlying crime be committed by some member of the conspiracy” and that the specific intent included the intent to send patients he knew to be federally insured to Forest Park. Anderson is inapposite. It is not an AKS case.[41] Finally, Henry admits to sending DOL/FECA patients to Forest Park. His only argument is that he did not know they were federally insured for purposes of the AKS. But there is sufficient evidence in the record that, because he was a licensed DOL/FECA provider, Henry knew that FECA was a federal program. Even if the Government were required to prove that Henry knew he was sending federal patients to Forest Park and that DOL/FECA was a federal program, there is sufficient evidence supporting both. D. Jacob Jacob argues that his conspiracy conviction cannot stand because he did not knowingly join the conspiracy. He claims that he had no knowledge that the payments Forest Park made to Adelaide were for referrals. Jacob’s argument fails under the weight of evidence in the record from which the jury could conclude that he knew exactly what was transpiring. Beauchamp testified that Jacob formed Adelaide specifically to be a pass-through entity for his referral program. Jacob acknowledges that paid patient referrals are illegal. Smith testified that she believed Jacob knew that the payments were for referrals. There are numerous emails corroborating this testimony. Jacob has no response to this evidence other than a claim that it is “speculative and inferential,” but that does not mean that there is not sufficient evidence for the jury to find him guilty. Further, he relies on Forest Park’s representation to him that the money was simply for marketing, as well as its representation to him that such marketing agreements were legal. This reliance ignores the evidence that Jacob was in on the conspiracy from the beginning. Forest Park certainly laid a paper trail to cover its tracks, but “it was within the sole province of the jury as the fact finder to . . . choose among reasonable constructions of evidence.”[42] E. Shah Shah’s argument fails for the same reason as the other surgeons’ (Won, Rimlawi, and Henry). Shah admits that his payments from Adelaide were for patient referrals. His only argument is that (1) there is no evidence that he knew accepting those payments was unlawful, and that (2) even if he did, there is no evidence that he knew DOL was subject to the AKS. As to his first argument, there is sufficient of evidence in the record from which a juror could infer that Shah, as a medical professional, knew taking money for patient referrals was unlawful. During cross-examination, Shah’s codefendant Rimlawi agreed that “taking money for patients is wrong” and testified, “I know I can’t take money for patients.” Several other witnesses testified likewise. As to his second argument, it fails for the same reasons Won’s and Henry’s argument fails. As noted in Part II(C) (Henry), even if the government had to prove that Shah knew his patients were federally insured and that DOL/FECA fell under the AKS umbrella, there is sufficient evidence in the record from which the jury could infer both. F. Forrest Forrest claims that there was insufficient evidence to sustain her conviction because nothing proved that she knew her involvement was unlawful. She claims that she thought the money was for preauthorization services. But the evidence supports the opposite inference. For one, in an email exchange between Forrest and Smith, Forrest asks, “How do the commissions work? I am on commission for a percentage of the surgeries that I send over. (just mine).” Smith replied that that was correct and requested that Forrest send over “an invoice for $10k.” At trial, Smith testified that Forrest was being paid for the referrals. Smith was asked, “[W]as it a service [Forrest] was paid for?” She responded, “To me it was the — the surgeries that were done.” Beauchamp’s testimony further cements that Forrest knew she was being paid for patient referrals, not preauthorization services. Beauchamp was asked, “Were you paying Ms. Forrest for preauthorization services, or were you paying her for surgical referrals?” He responded, “I was paying her for the surgical referrals, her surgical referrals.” Forrest further argues that the AKS does not apply to her because she is not a physician and she lacked “control over . . . physicians,” but the text of the statute is not so limited. It applies to “[w]hoever . . . solicits or receives any remuneration . . . in return for referring an individual.”[43] Forrest has no answer to this. And our caselaw makes clear that the AKS is not limited to those with “formal authority to effect the desired referral.”[44] It is enough that “remuneration [be] paid with certain illegal ends in mind.”[45] There is sufficient evidence in the record that Forrest was experienced in the healthcare field and that it was well-known in the healthcare industry that taking money in exchange for patient referrals was wrong. III Next, we turn to the substantive convictions. Jacob, Shah, and Forrest were convicted of violating the AKS. They challenge the sufficiency of the evidence supporting their convictions. A. Jacob Jacob argues that under the Government’s theory of the case, he was to be paid 10% of the kickback and that there is insufficient evidence to sustain his conviction because the checks the Government produced do not represent the theorized 10% kickback, nor can they be tied to individual patients. He also argues that he never induced Shah to steer patients to Forest Park because Shah gave the patients a choice of hospital. The Government counters that just because the checks do not equal 10% of the federal reimbursement does not mean they were not bribes. The Government also points to numerous emails detailing Shah’s complaints that he was indeed shorted his 10% and that Jacob questioned how accurate the tracking and payments were. Shah emailed Jacob: “10% was the number told to me by you and alan [Beauchamp].” Just because the math did not quite compute does not mean that the checks were not bribes. Based on these emails, the tracking sheets, and witness testimony from Beauchamp, the jury could have reasonably inferred that the checks were inducements or payments for referred patients in violation of the AKS. Jacob’s argument that the Government produced no evidence that the checks could be tied to the individual patients fares no better. At a minimum, the jury could have reasonably concluded that the checks Jacob and Shah received were for the patients Shah brought in on a monthly basis. There are numerous emails between the two men that demonstrate this knowledge— Shah complained to Jacob about being shorted month-to-month. Smith’s tracking sheets also track referrals and surgeries by month. Beauchamp’s testimony also established that payment was made on a monthly basis. There was sufficient evidence from which the jury could conclude that the checks supporting conviction were for patient referrals. Finally, Jacob’s contention that Shah never induced patients to go to Forest Park fails. Several witnesses said that Shah “gave [them] a choice” of clinic, but they all ended up at Forest Park. The jury chose to believe the Government over Shah, Jacob, and their witnesses. “We will not second guess the jury in its choice of which witnesses to believe.”[46] B. Shah In challenging the sufficiency of the evidence for the substantive AKS counts, Shah reiterates his arguments as to the lack of criminal intent for the conspiracy count. He also adopts by reference Jacob’s arguments as to the sufficiency of the evidence for the substantive AKS counts. Shah’s arguments fail for the same reasons as those discussed supra Part II(E) and Part III(A). C. Forrest Forrest’s arguments also fail. She reiterates her argument discussed above in Part II(F), contending that the fact she was not the patient’s doctor somehow excuses any inducement, but that argument fails for the reasons stated above. She also argues, like Jacob, that the Government could not tie the checks to her conduct. But the tracking sheets of Smith clearly tie Forrest to the patient, month of surgery, and check. The jury had sufficient evidence on which to convict. IV Burt and Henry challenge their Travel Act convictions, but there is enough evidence to convict each of them. The Travel Act prohibits the use of a “facility in interstate . . . commerce with [the] intent to . . . distribute the proceeds of an[] unlawful activity; or . . . otherwise . . . facilitate . . . an[] unlawful activity.”[47] To convict, the Government must prove that the defendant used facilities of interstate commerce with the specific intent to engage in or facilitate an unlawful activity in furtherance of a criminal enterprise.[48] The Supreme Court long ago recognized that the unlawful activity that predicates a Travel Act conviction may be commercial bribery in violation of a state statute, and it even cited the Texas statute at issue here as an example.[49] Further, this court has long held that a state statute serves merely to define the “unlawful conduct” required in the Travel Act and that there “is no need to prove a violation of the state law as an essential element of the federal crime.”[50] The state law at issue here is the Texas Commercial Bribery Statute (TCBS). The statute provides that it is a state felony for a physician to “intentionally or knowingly solicit[], accept[], or agree[] to accept any benefit from another person on agreement or understanding that the benefit will influence the conduct of the [physician] in relation to the affairs of his beneficiary.”[51] A. Burt Burt challenges his conviction on the ground that he was convicted on an aiding-and-abetting theory but that the physician he aided was acquitted. He argues that the TCBS would not support his conviction. He asserts there was no “unlawful conduct” for purposes of the Travel Act. The Government contends that the ultimate acquittal of the principal does not matter under Texas law and that federal law does not draw a distinction between principals and aiders and abettors. The Government is correct that federal law draws no distinction between principals and aiders or abettors.[52] But, more importantly, the Government is correct about the TCBS. Burt could still be found guilty of a violation of the TCBS even if his fiduciary physician was acquitted. This is because the TCBS criminalizes not only the fiduciary’s taking of the bribe, but also “offer[ing], confer[ring], or agree[ing] to confer any benefit the acceptance of which is an offense under [the statute].”[53] The Government produced evidence that Burt handled bribe money and at least offered it to if not conferred it on the physicians in question.[54] Because of this unlawful conduct, the fact that a physician was acquitted means nothing for purposes of Burt’s Travel Act conviction. Burt relies on United States v. Armstrong[55] for the proposition that he cannot be held liable when the principal was acquitted. But Armstrong is inapposite because the court there held that there was insufficient evidence to support the conviction, not that the defendant could not be convicted if the principal was acquitted.[56] Here, it does not matter if the physician was acquitted because there could still be sufficient evidence in the record that Burt “offer[ed]” a benefit in violation of the TCBS regardless of whether any physician accepted it.[57] B. Henry Henry was convicted of a violation of the Travel Act because commercial-bribery proceeds were moved via the internet from Forest Park into a bank account controlled by a pass-through entity and from there to Henry. He argues that he cannot be convicted because the Government failed to prove that a facility of interstate commerce was used or that Henry used such a facility. Specifically, he argues that the interstate passage of a check is too tangential to confer federal jurisdiction. He also argues that the Government could not prove any subsequent overt act on his part.[58] The Government responds that Henry relies far too heavily on inapposite, pre-internet caselaw and that it is now well established that the passage of a check via the internet is a use of the facilities of interstate commerce. This is true even for wholly intrastate transfers.[59] The Government has the better of the two arguments here. This court’s caselaw is clear that the use of the internet provides the interstate hook necessary for jurisdiction.[60] Henry’s out-of-circuit cases, predating this court’s more recent published decisions, are distinguishable and do not control the outcome here. Henry argues there is no evidence that the check traveled via the internet or that he personally used a facility of interstate commerce. It is undisputed that $30,000 was credited to Henry’s bank account, but he says that the bank employee who testified as to the interstate workings of the bank put forward hearsay when she said the check traveled through Illinois. He also argues that the Government put on no evidence that Henry had actually used a facility of interstate commerce. To the extent that the bank witness’s testimony that the check was cleared in Illinois was hearsay, it is irrelevant because all that is required under the Act is the use of an interstate facility—even if the entire transaction remained within the state.[61] Here, the check was indisputably routed over computer networks before clearing Henry’s bank account. As to Henry’s second point, that the Government cannot point to his actual use of interstate commerce facilities, the Government responds that he “caused the use of such facilities,” and that specific knowledge about the use of interstate facilities is “legally irrelevant” because the “words of § 1952 do not require specific knowledge of the use of interstate facilities.”[62] We have held that  “[t]here is no requirement that the defendant either have knowledge of the use of interstate facilities or specifically intend to use” them.[63] The jury could have inferred use of interstate facilities by the fact that the funds Henry received were transferred via electronic routing over computer networks. Finally, Henry challenges the evidence of a subsequent act. He contends that the government put forward no proof that he actually cashed the check. It is undisputed, however, that Henry received a $30,000 check from the pass-through entity and that the money subsequently was credited to Henry’s bank account. Henry’s only response is that there was no direct evidence that he deposited that money. But there is nothing in this court’s caselaw that requires such strict evidence of a subsequent act, and other circuits have held that “mere acceptance of the [bribe] money” is a sufficient overt act.[64] Further, there appears to have been no argument that someone other than Henry deposited the money. “[T]he relevant question is whether, after viewing the evidence in the light most favorable to the prosecution, any rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt.”[65] In the light most favorable to the prosecution, the jury could have found that Henry deposited the check. At the very least, the jury could have found that he accepted the bribe. V Next, Henry and Burt challenge their money laundering convictions. Henry and Burt were charged with conspiracy to commit concealment money laundering under 18 U.S.C. § 1956(a)(1)(B)(i), and there is sufficient evidence to show that they agreed to commit money laundering and that they joined the agreement knowing its purpose and with the intent to further it.[66] To prove the charge, the Government had to establish that the men conspired to “conduct a financial transaction with proceeds of a specified illegal activity . . . with the knowledge that the transaction’s design was to conceal or disguise the source of the proceeds.”[67] The predicate unlawful activity that produced illegal proceeds was the Travel Act violation discussed above. “Conspiracy to commit money laundering does not require that the defendant know exactly what ‘unlawful activity’ generated the proceeds.”[68] The defendant merely must know “that the transaction involve[d] profits of unlawful activity.”[69] The Government argues that it produced sufficient evidence that Henry and Jacob joined with Burt in a conspiracy to commit money laundering primarily through the testimony of Beauchamp. The Government points to the testimony of Beauchamp to argue that Burt was a mastermind of the operation alongside Beauchamp and that he worked with Jacob and Jacob’s company, Adelaide, to disburse illegal proceeds. The Government argues that Burt did the same with Henry, also based on Beauchamp’s testimony. The proceeds came from the Travel Act convictions, discussed above, which were predicated on bribery under the TCBS. The men concealed the illegal nature of the proceeds that Forest Park made on the bought surgeries by passing it through Adelaide and another entity, NRG, under consulting and marketing contracts. Beauchamp testified that the contracts were a sham and that both Burt and Henry knew it. Henry was instrumental in conceiving the idea of using NRG to funnel the proceeds to him. Henry counters that the Government produced insufficient evidence to prove a Travel Act violation and therefore could not prove a conspiracy to conceal the proceeds of that unproven Travel Act violation. Similarly, Burt argues that the evidence was insufficient to prove that the proceeds resulted from Travel Act violations. The Government responds, citing this court’s caselaw, that it “[is] not required to prove that [the defendants] actually committed the substantive offense[] of . . . money laundering” because this is a conspiracy charge.[70] The Government needed to prove only that the two men entered into an agreement to commit money laundering, that is, to conceal the illegal origin of ill-gotten proceeds,[71] and that they intended to carry it out.[72] The Government has met this burden through the testimony of Beauchamp who testified as to his relationship with Burt and the dealings between them and Jacob in creating Adelaide to funnel money to the surgeons under the guise of sham consulting contracts. Beauchamp testified as to the same with regard to Henry and NRG. A reasonable juror could have found conspiracy to commit money laundering on these facts. VI Won and Shah argue that the evidence proved several conspiracies, at odds with the indictment which alleged only one. Henry also raises this argument.[73] Forrest adopts this argument by reference.[74] Their argument fails. This court will affirm a “jury’s finding that the government proved a single conspiracy unless the evidence and all reasonable inferences, examined in the light most favorable to the government, would preclude reasonable jurors from finding a single conspiracy beyond a reasonable doubt.”[75] Even then, this court will only reverse if it finds prejudice.[76] The surgeons rely on several out-of-circuit cases to establish that the trial strayed from the indictment. Those cases lean heavily on wheel and chain models of conspiracies that have been firmly rejected by this circuit.[77] Their argument is that, at most, the Government attempted to establish several separate conspiracies rather than the one in the indictment. But this court does not use wheel and chain analogies to determine whether there is a single conspiracy. Rather, we look to “(1) the existence of a common goal; (2) the nature of the scheme; and (3) the overlapping of the participants in the various dealings.”[78] The surgeons fail to engage in this analysis, and even if they had, they would be unsuccessful. As to the first prong, this court interprets the “existence of a common goal” broadly.[79] A common pursuit of personal gain is sufficient, and that was unquestionably the goal of the conspiracy.[80] Second, as to the nature of the scheme, if the “activities of one aspect of the scheme are necessary or advantageous to the success of another aspect” then that supports a finding of a single conspiracy.[81] Here, although each surgeon was responsible for referring his own patients, his individual activities were advantageous to the success of the whole enterprise because Forest Park used that revenue to pay the pass-through entities as well as the surgeon. Moreover, the surgeons were necessary to “another aspect” of the conspiracy—unindicted non-surgeon bribe recipients. These non-surgeon bribe recipients referred patients to the surgeons who then passed them on to Forest Park. These non-surgeon recipients needed the surgeons to send those patients to Forest Park in order for the non-surgeons to receive payment from the conspiracy. Finally, regarding the overlapping of participants, this court finds that “[a] single conspiracy exists where a ‘key man’ is involved in and directs illegal activities, while various combinations of other participants exert individual efforts toward a common goal.”[82] That is the case here. Beauchamp, Toussaint, and Barker were the “key men.” They used Burt and Jacob to run the day-to-day operations, and they used the surgeons and Forrest to recruit patients all for the common goal of making money. In arguing otherwise, the surgeons cite Kotteakos v. United States,[83] which involved several separate conspiracies, but Kotteakos is easily distinguishable. In that case, “[t]here was no drawing of all together in a single, over-all, comprehensive plan.”[84] Even assuming no rational jury could have found a single conspiracy, the surgeons fail to show that this error “prejudiced [their] substantial rights.”[85] Henry and Forrest do not raise this point at all. Won and Shah address it only briefly and fail to provide any record citations to support the proposition that “clear, specific, and compelling prejudice” resulted in an unfair trial.[86] They argue that there was a great disparity in the quantity of evidence specific to them, but this court has held that quantitative disparities alone do not prove prejudice.[87] VII Henry argues that the TCBS is not a valid predicate offense to support a Travel Act conviction because it has been preempted by the Texas Solicitation of Patients Act (TSPA).[88] He first raised this argument in his motion to dismiss the indictment and repeats it on appeal. Henry’s argument is that these two statutes are in pari materia, meaning they “deal with the same general subject, have the same general purpose, or relate to the same person or thing or class of persons and things.”[89] According to Henry, the TSPA, as the more recent of the two, supplants the TCBS. We review this question of law de novo.[90] The Travel Act “aims to deny those engaged in a criminal business enterprise access to channels of interstate commerce.”[91] It provides, inter alia, that “[w]hoever . . . uses . . . any facility in interstate or foreign commerce, with intent to . . . distribute the proceeds of any unlawful activity[] or . . . otherwise promote, manage, establish, carry on, or facilitate the promotion, management, establishment, or carrying on, of any unlawful activity” may be fined or imprisoned.[92] The Supreme Court, citing the Texas statute as an example, has recognized that the unlawful activity that predicates a Travel Act conviction may be commercial bribery in violation of a state statute.[93] Henry does not contest this; rather, he argues that the TCBS has been supplanted by the TSPA by way of in pari materia. When two statutes are in pari materia, Texas law dictates that they should be harmonized.[94] The laws “should be construed together, and both given effect, if possible.”[95] It is only when the statutes “irreconcilabl[y] conflict[]” that “ the more specific statute controls.”[96] Henry argues that the two statutes conflict in that the TSPA incorporates the AKS safe harbor provisions whereas the TCBS does not.[97] In order for the TCBS and TSPA to conflict, conduct unlawful under the TCBS must fall within a defense provided for in the TSPA. Because the two statutes criminalize nearly identical conduct, the only way for this to be the case is if something in the safe harbor provisions incorporated into the TSPA would prevent conviction that otherwise would be proper under the TCBS.[98] There are twelve exceptions to the AKS found in the safe harbor provision.[99] Henry addresses none of them. Henry has forfeited this argument by failing to brief it adequately.[100] Even assuming the statutes are in pari materia, Henry cites no authority for why the latter would supplant the former. As discussed above, Texas law requires that the statutes be harmonized if possible.[101] If both cannot be given effect, then the more specific statute would control.[102] Moreover, the Supreme Court has long recognized that violation of state commercial bribery statutes is a valid predicate for Travel Act convictions,[103] and this court has long held that a state statute serves merely to define the “unlawful conduct” required in the Travel Act.[104] There “is no need to prove a violation of the state law as an essential element of the federal crime.”[105] We decline to depart from this long-settled precedent. VIII Won argues that the district court erred in denying his motion to dismiss the indictment for a violation of the Speedy Trial Act (STA).[106] The district court did not err in denying Won’s motion because Won consented to a continuance encompassing most of the delay he now challenges. In May 2017, the parties requested and the court granted an “ends-of-justice” continuance through January 2018.[107] In November 2017, Judge Fitzwater (the original judge assigned to this case) announced he was taking senior status and his intention to transfer this case to another judge. Because that process would take at least several months to complete, he vacated the late January 2018 trial date. Won did not object at that time. In late January, Chief Justice Roberts assigned Judge Zouhary to the case, and again without any objection from the defendants, Judge Zouhary set trial for early 2019. It was not until October 2018 that Won objected to any delay. The STA “‘generally requires a criminal defendant’s trial to start within 70 days of his indictment or his appearance before a judicial officer,’ whichever date last occurs.”[108] But the STA includes a “long and detailed list of periods of delay that are excluded” from the 70-day window.[109] Relevant to this appeal, the STA excludes delay resulting from a continuance on the basis that the ends of justice outweigh the interest of the public and the defendant in a speedy trial.[110] Won does not appear to dispute that the May 2017 continuance through January 2018 was an ends-of-justice continuance. Nor is it disputed that motion filings in early February 2018 tolled the 70-day clock. His only argument for an STA violation is that the November 2017 order vacating the January trial date reset the STA clock and that there are more than 70 non-excludable days between November 17, 2017, and February 2018 when the filing of motions stopped the clock. We review the district court’s factual findings for clear error and its legal conclusions de novo.[111] This court has held that defendants are precluded from challenging any delay to which they have consented.[112] Won consented to the May 2017 ends-of-justice continuance setting the trial date for no earlier than January 2018. He cannot now object to any delay between November 2017 and January 2018 to which he has already consented.[113] He cites no authority to support his argument that Judge Fitzwater’s November order vacating the January trial date has any effect on his ability to challenge a delay to which he had already consented. Nor does hesupport his argument that the November order restarted the 70-day clock, and there is caselaw to support the proposition that the November order did not restart the clock. In United States v. Bieganowski,[114] for example, this court suggested that an ends-of-justice continuance excluded all the days of the continuance from STA calculations even though a later act arguably restarted the clock.[115] In Bieganowski, the court granted an ends-of-justice continuance until August 23.[116] On August 12, the court granted another continuance, this one until November.[117] The court also granted a third continuance in September.[118] The first and third continuances satisfied the requirements of the STA.[119] The second continuance arguably did not, but this court declined to answer the question of whether it did because the third continuance met the requirements of the STA.[120] Key to the court’s analysis was the fact that only 10 days passed between the end of the first continuance and the beginning of the third.[121] The questionable second continuance was granted prior to the end of the first one, yet this court used the end of that first continuance as the point at which the STA would restart assuming the second continuance was contrary to the STA. In other words, the court’s actions prior to the end of the first continuance had no effect on the STA calculations because the parties had consented to the entirety of that first continuance. So too here. It is undisputed that Won consented to the May 2017 continuance through January 2018. It is also undisputed that the 70-day clock as tolled on February 3, 2018. Won cannot point to more than 70  non-excluded days. IX Won next argues that the district court violated the Court Reporter’s Act[122] (CRA) when it went off the record 46 times during the 29-day trial. Jacob adopts this argument specifically as to the court’s failure to record the charge conference. But whatever gaps exist in the record of this case do not amount to a violation of the CRA. The CRA provides that “[e]ach session of the court” in a criminal proceeding “shall be recorded verbatim by shorthand, mechanical means, electronic sound recording, or any other method.”[123] In cases, as here, where appellate counsel was not trial counsel, a CRA violation occurs when “a substantial and significant portion of the record” is missing such that “even the most careful consideration of the available transcript will not permit [this court] to discern whether reversible error occurred.”[124] But this court has long held that “a gapless transcription of a trial is not required.”[125] “We have not found reversible error even when a transcript was missing seventy-two bench conferences.”[126] “[A] merely technically incomplete record” is not error.[127] Won argues that the 46 missing bench conferences robbed his appellate counsel of the rationale for various district court rulings, especially the exclusion of some of Ford’s testimony and several exhibits. Without that rationale, Won argues, he faces substantial prejudice because he cannot mount an appeal. The first question presented to this court is the standard of review. Won claims that he raised his CRA argument to the district court in a table of evidentiary rulings he filed mid-trial and that he presents a question of law reviewed de novo. This table memorialized Won’s objections to various rulings, but it did not raise the CRA directly. The closest it came to the CRA was mentioning in a footnote that “many of the evidentiary rulings regarding trial exhibits in this case occur[red] off of the record.” Won then explains that he filed the list to “reflect[] the current status of the trial exhibits admitted and excluded” including “Dr. Won’s objections.” Won neither raised the CRA nor objected to the court’s procedure. Accordingly, we agree with the Government and review the potential violation for plain error.[128] Won must show that the error was “plain,” “affected [his] substantial rights,” and “seriously affected the fairness” of his trial.[129] Won cannot show plain error. This court has only recognized CRA violations for truly egregious omissions like an absence from the record of voir dire, opening statements, closing arguments, or even an entire transcript.[130] Won does not point this court to any case in which the court found reversible error for off-the-record bench conferences, especially when objections were later memorialized. The Government, on the other hand, points this court to a litany of cases in which the court has not found reversible error even in the face of several dozen more missing conferences than at issue here.[131] The district court did not plainly err. X Burt argues that the district court erred by finding that he had breached his pre-trial proffer agreement with the Government. We hold that the district court did not commit clear error in determining that Burt offered evidence inconsistent with his proffer and that this constituted a breach of his agreement. Well before trial, Burt engaged in a proffer agreement with the Office of the Inspector General. He agreed to tell the truth about Forest Park in exchange for the Government not using his statements against him. The agreement, which is interpreted according to the general principles of contract law,[132] stated that the Government would not use Burt’s statements against him in the Government’s case-in-chief “except . . . for statements outside the proffer that are inconsistent” with the proffer. In a later paragraph, the agreement makes clear that if Burt or his attorney elicited “arguments that are inconsistent with [the proffer,] . . . [then the Government] may use proffer information to rebut or refute the inconsistencies.” In his proffer interview, Burt stated that “[y]ou don’t entice doctors because that would be against the law” and that he realized from the beginning that the $600,000 check Beauchamp paid to Adelaide was for kickbacks. In pre-trial filings, Burt argued that he did not know that the checks were for kickbacks and that he was generally unaware of impropriety. The Government objected, claiming that he had breached his proffer agreement. The court held an evidentiary hearing and agreed that Burt had breached the proffer and that the remedy, according to the agreement, was for the Government to be able to rebut any breach statements that Burt elicited at trial. At trial, Burt’s attorney cross-examined Forest Park’s former controller, David Wheeler who had testified to various improprieties at Forest Park. To impeach Wheeler, Burt used a representation letter that not only Wheeler but also Burt had signed. The letter generally attested that none of the signatories had knowledge of fraud within the hospital. Burt’s attorney made use of a projector for this part of his cross examination, blowing up the representation letter on the screen for the jury. The attorney made repeated references to the signatures depicted on the screen, blew up the signature page until it was quite large, and told the jury to “look at the signatures” while eliciting testimony from Wheeler that those signatures, including Burt’s, attested to the fact that there was no fraud or impropriety. The Government renewed its objection that Burt had breached the proffer agreement. It argued, as it does on appeal, that the testimony Burt’s attorney elicited from Wheeler that the signatures meant that no one knew of any fraud directly contradicted Burt’s earlier statement that he knew about the kickbacks all along. The court agreed, concluding that Burt had breached the agreement and that the Government was entitled to rebut Burt’s assertion that he had no knowledge of fraud. The parties disagreed as to how. After a lengthy discussion with the parties, the court settled on a remedy whereby the judge would read an agreed-to statement to the jury. That statement reads as follows: Defendant Mac Burt made statements in June 2016 to Casey England, an agent with the Office of Inspector General. You may have heard those initials OIG during the trial. Those statements were during a voluntary interview where he was represented by legal counsel. The interview, consistent with Department of Justice policy, was not taped. The agent took notes. Those notes include a statement by Defendant Burt that he realized from the very beginning that the $600,000 check Beauchamp requested from Forest Park to be paid to Adelaide was for doctor kickbacks. You may consider this evidence as to Defendant Burt. Burt first claims that he did not breach the agreement because the testimony was merely used to impeach Wheeler. Second, Burt claims that the district court misinterpreted the proffer agreement by allowing the Government to rebut any inconsistency during its case-in-chief. Third, Burt argues that the court’s remedy was error. Finally, Burt argues that any error was not harmless. A The district court’s finding of breach is reviewed for clear error.[133] We review de novo whether, under those facts, the agreement was in fact breached.[134] A factual finding is clearly erroneous only if it is implausible “in light of the record as a whole.”[135] The court referenced the testimony of Wheeler as well as the proffer agreement and found them to be inconsistent. We agree. Burt bargained with the Government to tell the truth in his proffer and to not make inconsistent statements at trial. The agreement was explicit that statements Burt elicited would count as inconsistent. Burt was on notice for several months that he was violating the proffer every time he tried to argue that he had no knowledge of any fraud or impropriety, yet at trial he elicited testimony contrary to his proffer. Wheeler’s testimony that the representation letter was an attestation of no impropriety, when combined with Burt’s attorney’s focus on the signature page containing Burt’s signature, leads to a not clearly erroneous conclusion that Burt was acting inconsistently with his earlier statement that he had knowledge of wrongdoing. Burt’s argument to the contrary is not persuasive. He argues that he was merely impeaching Wheeler, and he relies heavily on a Seventh Circuit case for the proposition that defendants ought to be given broad leeway to impeach government witnesses even while under the stricture of a proffer agreement.[136] But United States v. Krilich[137] does more to hurt Burt’s argument than help it. There, the Seventh Circuit affirmed the district court’s determination that the defendant breached the proffer because he was not merely impeaching the witness; rather, his counsel was eliciting statements “inconsistent with the proffer.”[138] So too here. B Burt also argues that the district court erred by not harmonizing the agreement’s provision protecting him from the Government’s use of any statement in its case-in-chief with the provision allowing rebuttal evidence. We review the district court’s interpretation of the proffer agreement de novo.[139] Burt’s argument fails on its face. The court expressly explained the two provisions’ interaction, concluding that the latter provided the Government with a rebuttal remedy should Burt breach the agreement not to make inconsistent statements. We agree that the agreement unambiguously creates that remedy. Nor is Burt correct to argue that the rebuttal was precluded from taking place in the Government’s case-in-chief. It is true that paragraph 3 of the agreement explains that the Government would not use Burt’s proffer against him in its case-in-chief, but the agreement included an express exception for inconsistent statements. Paragraph 7 clearly provides that the remedy is rebuttal. Moreover, this court has recognized that “rebuttal waiver[s] might be worded so broadly as to allow admission of plea statements in the government’s case-in-chief.”[140] If rebuttal could not take place during the case-in-chief, the Government might never get an opportunity to hold defendants accountable for breaching the agreement because defendants can choose not to present a case at all.[141] C Burt argues that the district court’s remedy of reading a statement to the jury prejudiced him, and he urges this court to review that decision for abuse of discretion. This court does not appear to have addressed a standard of review for the remedy chosen by the district court, nor does the Government in its brief. We have suggested, however, that we would review the admission of plea negotiation evidence for abuse of discretion.[142] We reasoned that an objection to the admission of such evidence would be no different than an objection to any other evidence and that the same abuse of discretion standard should apply.[143] Other circuits have approached breaches of plea agreements in accordance with contract principles, reasoning that “[i]t is for the district court to decide what remedy is appropriate.”[144] We adopt the abuse of discretion standard here. Burt argues that the proper remedy should have been either: (1) an instruction that Wheeler’s testimony could only be considered as to Wheeler’s knowledge and beliefs and not Burt’s; or (2) an opportunity to cross-examine the agent who interviewed Burt. But in doing so, Burt essentially asks this court to strike a different balance than that of the district court. That is not our role in reviewing for abuse of discretion. “A trial court abuses its discretion when its ruling is based on an erroneous view of the law or a clearly erroneous assessment of the evidence.”[145] As explained above, the district court was correct in determining that Burt elicited inconsistent statements and in concluding that they amounted to a breach of his proffer agreement. It is hard to see how reading the statement was an abuse of discretion. Burt falls back on the argument that the district court’s decision to read the statement as opposed to allow Burt to cross-examine the agent who interviewed him violated his due process rights.[146] He cites little in the way of elaboration, and it does not appear that he raised this argument in the district court. What little analysis he provides is simply a rehash of his earlier arguments that he did not breach the agreement and an objection that the prosecutor characterized the statement as a “confession” during closing arguments. This argument is forfeited for lack of adequate briefing.[147] Nor does the (limited) argument Burt makes with regard to a violation of the Confrontation Clause affect this analysis. Burt waived any Confrontation Clause challenge at trial. Even if he had not, he has not adequately briefed it here and we would deem it forfeited.[148] Further, even assuming the court erred, any error was harmless given the other, substantial evidence against Burt, including testimony from numerous witnesses that he did in fact know what was going on from the beginning and that the money was for bribes and illegal kickbacks. XI Jacob, Rimlawi, Won, and Henry all argue that the court erred by reading a portion[149] of Burt’s proffer into the record. The defendants argue that this was Bruton error. Shah and Forrest adopt the arguments of their codefendants. Rimlawi further argues that the court erred in allowing the prosecution to cross-examine him with the proffer.[150] Because the proffer “could only be linked [to the defendants] through additional evidentiary material,” there was no Bruton error.[151] Rimlawi’s argument, however, fares better. Assuming the district court erred by cross-examining Rimlawi with the proffer, that error was harmless. We will address the threshold challenge to the admission of the proffer raised by Jacob and the physicians below. We will then address Rimlawi’s challenge to the proffer’s use during his cross-examination. But first, the defendants challenge the exact wording of the court’s limiting instruction. They urge this court to reverse because the court limited the use of the proffer “as to Defendant Burt” and not as to Burt only. The omission of “only” in the limiting instruction, they argue, is reversible error. The parties have not provided any caselaw on point to support their assertion, nor have we found any. We are not convinced that the omission of “only” is reversible error. We may safely assume “the almost invariable assumption . . . that jurors follow their instructions.”[152] The instruction given was that the jury may consider the proffer “as to Defendant Burt.” The “only” is implied. Additionally, any error in the instruction was harmless given the weight of evidence against all of the defendants. A In Bruton, the Supreme Court held that the admission of a non-testifying codefendant’s statements may violate a testifying codefendant’s Sixth Amendment right to confront his accusers.[153] But “[o]rdinarily, a witness whose testimony is introduced at a joint trial is not considered to be a witness ‘against’ a defendant if the jury is instructed to consider that testimony only against a codefendant.”[154] There is “a narrow exception to this principle.”[155] If the admitted testimony “facially incriminate[s]” the defendant, then the admission may violate the defendant’s Confrontation Clause rights even if the court gives a limiting instruction.[156] Further, although it is assumed that “jurors follow their instructions,”[157] the “prosecution [can] upend[] this assumption” by “clearly, directly, and repeatedly” using the non-testifying codefendants’ statements against a testifying codefendant.[158] Such use of a non-testifying codefendant’s statements is “a clear and obvious violation of a constitutional right that substantially affects the fairness of judicial proceedings” and is plainly erroneous.[159] The “key analytic factor” in deciding whether there is Bruton error is whether the admitted proffer “clearly refer[s]” to the other codefendants or whether it “ could only be linked through additional evidentiary material.”[160] If further linkage is required, then the proffer does not “facially implicate[]” the other physicians and it does not violate their Sixth Amendment rights.[161] We review constitutional challenges de novo, but we review the trial court’s  “evidentiary decisions on a Bruton issue . . . for abuse of discretion.”[162] Bruton errors are subject to harmless error analysis.[163] The only objectionable part of the court’s statement to the jury was that “[Burt] realized from the very beginning that the $600,000 check Beauchamp requested from Forest Park to be paid to Adelaide was for doctor kickbacks.” The physicians (Won, Rimlawi, and Henry) argue that the court’s use of “doctor” facially implicated them. Jacob argues that the reference to his company, Adelaide, is enough to facially implicate him. In United States v. Powell,[164] on which the defendants rely, this court held that the admission of a non-testifying codefendant’s statement to an investigator did not violate Bruton.[165] A husband (Powell) and his wife (Akin) transported cocaine together in their car. They were stopped by police and interviewed separately. Akin made several inculpating statements to investigators that the prosecution used at trial against Powell. The statements related to Akin’s knowledge that the car she was a passenger in was transporting crack cocaine.[166] Akin did not testify at trial. This court held that the admission of the statements “did not directly implicate” Powell despite the fact that it was well established that the two were in the car together.[167] The testimony concerned only Akin’s knowledge and actions— any relation to Powell had to be inferred. So too here. Although the proffer statement directly mentions “doctors” and Adelaide, further evidence is required to link Won, Rimlawi, and Henry to “doctors” and Jacob to “Adelaide.” Burt’s use of “doctors” could have referred to any number of physicians. The fact that the three defendants were on trial and also doctors does not mean that the use of “doctors” facially implicated them. The jury had to examine other evidence to determine whether those three doctors were indeed the doctors who had received kickbacks. All the proffer stands for directly is that Burt knew Beauchamp was paying physician kickbacks. The jury had to decide which physicians were receiving kickbacks. Likewise, although the statement directly refers to Adelaide, more is required to link Jacob to Adelaide. First, of course, would be evidence that Jacob operates Adelaide. Second, the jury would have to find that any illegal actions of Adelaide could be imputed to Jacob. More evidence was required to link Jacob to the illegal conduct for which he was eventually convicted. B Even if the admission of the proffer statement was not error, and we hold that it was not, that does not end the Bruton analysis. This court has recognized that while admission of a non-testifying defendant’s statement may not be erroneous if properly limited, that statement’s use against other defendants outside the limiting instruction may violate the Confrontation Clause.[168] Here, the district court limited the proffer’s use by instructing that the jury may consider the statement “as to Defendant Burt.” Nonetheless, the prosecution’s subsequent use of the proffer against Rimlawi may have been in error.[169] When Rimlawi took the stand in his own defense, the Government used the proffer against him directly. Rimlawi claimed that he “didn’t have any deal or side deal that was illegal or involved kickbacks.” The Government cross-examined him with the statements of several individuals who had testified that Rimlawi had in fact been “paid for patients.” The prosecutor listed 10 individuals who had testified that Rimlawi was involved in the kickback scheme. At the end of this list and as the eleventh individual to testify against Rimlawi, the Government briefly mentioned Burt’s proffer statement. Rimlawi claims that this admission violated the Confrontation Clause. In Powell, discussed above, this court determined that the admission of Akin’s statement was not Bruton error, but it held that the prosecution’s use of that statement to cross-examine Powell was erroneous.[170] The Government attempts to distinguish Powell, contrasting the extent of the cross-examination in that case versus here. It is true that the cross-examination in Powell focused more on the potentially violative statement than here—the prosecutor brought up Akin’s statement five times in a row.[171] But the rationale in Powell was that the prosecution “upended” the court’s limiting instruction when it used the statement “clearly, directly, and repeatedly” against Powell.[172] While the extent of the use of the proffer at issue here is less than in Powell (used once versus five times), it was “clearly” and “directly” used against Rimlawi. That use may violate Rimlawi’s constitutional right to confront his accusers. But even assuming without deciding that the admission of the statement in cross-examination was error, that error was harmless. “It is well established that a Bruton error may be considered harmless when, disregarding the co-defendant’s confession, there is otherwise ample evidence against the defendant.”[173] To find an error harmless, we must be convinced beyond a reasonable doubt that the error was in fact harmless in light of the other evidence presented at trial.[174] We will not find a Bruton error harmless if there is “a reasonable probability that the defendants would be acquitted.”[175] In Powell, the court held that even though the admission during cross-examination was plain error, the error was harmless and the conviction could stand because of the weight of the other evidence against Powell.[176] So too here. As evident in the exchange at issue for Rimlawi, no fewer than 10 other individuals implicated him in the kickback scheme. Just as Powell was caught driving “a car loaded with crack cocaine packaged for sale,” a mountain of other evidence inculpates Rimlawi.[177] As discussed above in Part II(B), Beauchamp testified that Forest Park paid for federally insured patients. Rimlawi admits to having federally insured patients. Smith’s kickback tracking sheets show that Rimlawi was credited with DOL/FECA insured patients, and Rimlawi does not contest that DOL/FECA patients are federal pay. XII Won, Rimlawi, and Shah argue that the district court abused its discretion in excluding various portions of two witnesses’ testimony: Theresa Ford and Bill Meier. The court did not abuse its discretion. The surgeons argue that the district court erred in excluding portions of Ford and Meier’s testimony along with a related email from Ford and certain billing invoices from Meier. The surgeons attempted to introduce this evidence to bolster their advice-of-counsel defense. The surgeons suggest now that neither attorney was able to testify at trial meaningfully, but that is not the case. Both attorneys testified at trial. The surgeons’ appeal focuses on three sets of excluded evidence: (1) an email Won wrote to Ford as well as testimony that Won told Ford that Forest Park did not accept federally insured patients; (2) Ford’s testimony regarding how common marketing schemes for physicians are and her opinion that Forest Park’s was legal; and (3) Meier’s testimony concerning the same. The court ruled that the attorneys could testify “as relevant to the state of mind of a defendant,” but they were not allowed to “be a mouthpiece for the defendant” or to “offer legal opinions.” The court did not allow the lawyer-witnesses “to make legal conclusions or opinions” with regard to central issues in the case. It excluded the evidence at issue on a variety of grounds. The district court found testimony about the legality of the marketing scheme to be irrelevant given that the marketing agreement was, on its face, legal and not at issue. It also excluded the testimony regarding the ultimate legality of the programs as legal conclusions by a lay witness. It concluded that some of the conversations between the surgeons and attorneys about whether the surgeons’ actions were legal were hearsay. Evidentiary rulings are reviewed for abuse of discretion.[178] The harmless error doctrine applies.[179] Irrelevant evidence is inadmissible.[180] So too is hearsay evidence that does not fall within an exception.[181] A lay witness’s opinion testimony is limited to opinions that are “based on the witness’s perception[,] helpful[,] . . . and not based on . . . specialized knowledge.”[182] A The district court did not abuse its discretion in excluding the Ford email because the only statements the surgeons object to are hearsay. The surgeons object to the exclusion of three statements: (1) a statement that the hospital “only accepts private commercial insurance,” and “do[es] not accept any federally funded programs and [has] no plans to do it in the future”; (2) a statement that Forest Park told Won it was not “participating in any federally funded program” or “affected by stark or anti-kickback issues”; and (3) a statement that Won “want[ed] to make sure we are compliant.” Federal Rule of Evidence 803(3) creates an exception to hearsay for statements concerning a declarant’s “then-existing state of mind” but not for “a statement of memory or belief to prove the fact remembered or believed.”[183] The Government argues that the first two statements listed above fall outside Rule 803(3) because they are statements of memory or belief offered to prove the fact remembered or believed. In each case, the surgeons seek admission of testimony that Forest Park was not connected to federally funded programs to prove the same. This court held, in nearly identical circumstances, that this is “the kind of statement of historical fact or belief that Rule 803(3) precludes.”[184] We see no reason for a different result here.[185] As for the third statement, Won’s only argument is that the statement was a verbal act and not hearsay. He does not raise Rule 803(3) with regard to that statement and has forfeited that argument.[186] The statement itself is not a verbal act within the meaning of the term because he sought to admit it for the truth of the matter asserted, i.e., that he sought compliance.[187] Shah raises a distinct challenge to the exclusion of this evidence. He asserts that its exclusion violated his Sixth Amendment right to present a complete advice-of-counsel defense. Even under de novo review, which would apply here,[188] Shah’s argument lacks merit. The right protected is to  “present a defense” in part by “present[ing] his own witnesses to establish a defense.”[189] Shah fails to address the fact that Ford did in fact testify about her relationship with Won and Rimlawi (she does not appear to have ever worked with Shah). Rimlawi’s attorney managed to ask about whether the surgeons sought compliance with all applicable laws during her allotted time to examine Ford. It is hard to see how Shah was not afforded the opportunity to present a defense. United States v. Garber[190] is not to the contrary. There, the defendant’s witness was prevented from testifying to the existence of a legal theory supporting the defense.[191] This court found error.[192] Here, on the other hand, the statements the district court excluded are simple, run-of-the-mill hearsay statements from Won. Ford was allowed to testify as to what she looked for in making sure marketing agreements were legal. Garber is inapposite. Rimlawi challenges the exclusion of his own testimony related to this same topic, i.e., his state of mind and advice-of-counsel defense. He argues the district court erred in excluding this testimony as hearsay because the testimony was offered not for the truth of the matter asserted, but rather to show his state of mind. We need not address whether this testimony was properly excluded. Even if the court abused its discretion in excluding this testimony, any error was harmless. The district court ruled that the attorneys could testify “as relevant to the state of mind of a defendant.” Accordingly, the district court allowed attorneys Ford and Meier to testify about the advice they gave the defendants. Attorney Meier even testified as to his conversations with Rimlawi in particular. As a result, Rimlawi’s testimony would have been duplicative of the attorneys’ testimony and would not have had an impact on the jury’s guilty verdict.[193] B Next, the surgeons argue that the district court erred by not allowing Ford to testify as to the categorization of healthcare programs and the legality of the marketing agreement she reviewed. The court did not err. First, Ford was not allowed to testify to the jury as to whether DOL/FECA is a federal healthcare program. But she was not qualified as an expert witness, and the surgeons did not establish that she had personal knowledge of the source of DOL funding. There is no abuse of discretion in precluding a lay witness from testifying as to something of which they lack personal knowledge.[194] Second, Ford was also not allowed to testify that comarketing agreements are common and that Forest Park’s actual arrangement was legal. But Ford was allowed to testify about comarketing in general and the marketing agreement Won had sent her in 2009 (which was not the one that ended up being the operative agreement between Won and Forest Park pass-through entities). The Government does not contest that the marketing agreement was facially legal. What mattered, the Government urges, is what the agreement did not say—that the physicians were accepting illegal kickbacks as part of this agreement. Ford did not have personal knowledge of these facts. She could not opine that the agreement Won and the pass-through entity reached and operated under was legal.[195] C Attorney Meier was also not allowed to testify as to the legality of the surgeons’ marketing agreements. For the same reasons as above, the district court did not err. XIII Next, Won and Rimlawi argue that the district court erred by denying their request for specific jury instructions on advice-of-counsel and good-faith defenses. Jacob and Shah argue that the district court erred by denying the good-faith instruction. Forrest adopts by reference arguments as to the denial of the good-faith instruction. The district court did not abuse its discretion in declining the defendants’ request for the two jury instructions. The good-faith instruction was covered by the jury instructions given. Won and Rimlawi were not entitled to an advice-of-counsel instruction because there was not a proper foundation for it in evidence. Won and Rimlawi appeal the district court’s denial of their request for specific jury instructions as to advice-of-counsel and good-faith defenses. Shah and Jacob appeal the denial of the good-faith instruction, and Jacob also appeals the district court’s ultimate instruction on willfulness because it “exceeded the circuit pattern.” We review the denial of a jury instruction under an “exceedingly deferential” abuse of discretion standard.[196] We afford district courts “substantial latitude in tailoring” their jury instructions so long as the instructions “fairly and adequately cover the issues presented.”[197] The district court abuses its discretion only if “(1) the requested instruction is substantively correct; (2) the requested instruction is not substantially covered in the charge given to the jury; and (3) it concerns an important point in the trial so that the failure to give it seriously impairs the defendant’s ability to effectively present a particular defense.”[198] A The defendants’ argument that the district court erred by denying their requested good-faith instruction fails because the jury instructions the court gave covered the good-faith instruction it denied. This court has held that “the omission of a good faith jury instruction is not an abuse of discretion if the defendant is able to present his good faith defense to the jury through, inter alia, witnesses, closing arguments, and the other jury instructions.”[199] Key among these other jury instructions are those related to “knowing” and “willful” conduct because good-faith reliance defenses depend on disproving knowing or willful elements of the crime.[200] In other words, so long as the defendants are able to present their good-faith defense within the existing jury instructions regarding “knowing” and “willful” conduct, there is no error. Here, the district court’s instructions concerning “knowing” and “willful” conduct are similar to those in United States v. Frame[201] and United States v. Davis.[202] Although unpublished, the analysis in Frame is informative. There, this court affirmed the denial of the jury instruction as to a good-faith defense because it was captured within the jury instructions actually given; the court held that the instructions made plain that the jury was required to acquit Frame if, “because of his good faith, he lacked specific intent.”[203] Likewise, in Davis this court affirmed the denial of a requested jury instruction as to good faith because “those concepts were adequately explained through the district court’s definitions of the terms ‘knowingly’ and ‘willfully.’”[204] The same result holds here. The district court instructed the jurors that the Government had to prove that the defendants acted knowingly, which it defined as “done voluntarily and intentionally and not because of mistake or accident.” It then defined “willfully” as an “act [that] was committed voluntarily and purposely with the specific intent to do something that the law forbids, that is to say, with the bad purpose either to disobey or disregard the law.”[205] These instructions make clear that the jury could not convict the surgeons if they found that they had acted without the specific intent to do something the law forbids, i.e., if they were acting in good faith.[206] In addition, Jacob’s argument that the district court’s willfulness instruction here “exceeded circuit pattern” is unsupported by caselaw and fails. B Won’s and Rimlawi’s argument that the district court erred by denying their requested advice-of-counsel instruction fails because they failed to establish the requisite evidentiary foundation. A court “may . . . refuse to give a requested instruct[ion] that lacks sufficient foundation in the evidence.”[207] An advice-of-counsel defense has four elements: (1) before taking action, the defendant in good faith sought the advice of an attorney; (2) for the purpose of securing advice on the lawfulness of potential future conduct; (3) gave a full and accurate report of all material facts; and (4) the defendant acted strictly in accordance with the attorney’s advice.[208] A successful advice-of-counsel defense negates willfulness by “creat[ing] (or perpetuat[ing]) an honest misunderstanding of one’s legal duties.”[209] Even assuming without deciding that the defendants can meet the first and second prongs of the test, they fail to meet the third and fourth. It is undisputed that Ford only billed 1.3 hours and did so preparing an agreement that ended up not being used. Further, neither Ford nor Meier was aware of the surgeons’ full dealings with the principals of Forest Park. They explicitly informed the surgeons that they should not accept kickbacks for patient referrals, yet that is exactly what the surgeons did. The surgeons do not satisfy the fourth prong of the defense as well. XIV Shah argues that the district court erred by not instructing the jury on multiple conspiracies and instead instructing it only as to a single conspiracy, as alleged in the indictment. Shah’s argument is counter to well-settled precedent. Shah failed to make his objection during trial, so plain error review applies.[210] Shah argues that he preserved the objection in a document of proposed instructions he filed before trial even began. But nowhere in his 145-page document does he note his “specific objection and the grounds for the objection” as required by Federal Rule of Criminal Procedure 30.[211] Plain error review applies. “[A] failure to instruct on multiple conspiracies generally does not constitute plain error.”[212] Shah cannot show error here because a lack of a multiple-conspiracies instruction did not prejudice his defense that he never conspired in the first place.[213] Won, Forrest, and Jacob attempt to adopt Shah’s argument here by reference. The Government argues that this argument cannot be adopted by reference because the analysis is too fact specific. Even assuming without deciding that Shah’s argument could be adopted by reference, any adoption would fail for the same reasons discussed above. XV Shah and Jacob raise myriad complaints about the prosecutors’ actions during closing argument. Forrest adopts these arguments by reference. Even assuming the prosecutors engaged in some misconduct during closing argument, the defendants have failed to establish that the misconduct affected their substantial rights. We review allegations of prosecutorial misconduct with a two-step analysis: first, we look to whether the prosecutor “made an improper remark”; if so, we analyze whether that remark affected the defendant’s “substantial rights.”[214] The defendants did not raise their objections at trial, so we review them for plain error.[215] Reversing a conviction “on the basis of a prosecutor’s remarks alone” is not a decision this court makes “lightly.”[216] “[T]he determinative question is whether the prosecutor’s remarks cast serious doubt on the correctness of the jury’s verdict.”[217] This is a “high bar.”[218] This court considers “the magnitude of the prejudicial effect,” “the efficacy” of any instructions, and “the strength of the evidence.”[219] Even if the surgeons can meet this high burden, this court retains discretion whether to reverse, “which we generally will not do unless the plain error seriously affected the fairness, integrity, or public reputation of the judicial proceeding.”[220] The alleged misconduct can be summarized as follows: improper vouching; personal attacks; misstatement of the evidence; telling jurors they are victims; faulting the defense’s choice to remain silent; and shifting the burden of proof. But most of the objected-to conduct is not objectionable when viewed in context. For example, Shah and Jacob argue that the prosecutor faulted the defense’s choice to remain silent, but when viewed in context, all of the statements relate to the paucity of the evidence the defense did put on to support their various defenses.[221] Similarly, the defendants’ objections as to burden-shifting fail for the same reason—the prosecutor’s statements referred to their lack of evidence for affirmative defenses.[222] Shah also argues that the prosecutors committed misconduct by telling the jurors that the jurors were victims and by making personal attacks against the defendants. These arguments carry more weight. The prosecutors referred to the effect the fraud had on the medical system in the United States, explaining to the jurors that “[t]here are a lot of victims in this case” and that “[t]he greed of the defendant[s] impacted us as a community.” Shah complains that this amounted to a “so-called ‘golden rule’ argument” because it urged the jury to put itself into the shoes of the victim.[223] Citing out-of-circuit precedent, Shah contends that such arguments are “universally condemned.”[224] He also argues that “invoking the individual pecuniary interests of jurors as taxpayers” is improper. In response, the Government points this court to United States v. Robichaux[225] for the proposition that the prosecutors were allowably “impress[ing] upon the jury the seriousness of the charges.”[226] There, this court found no error in the statement that “Louisiana citizens and all those who seek to purchase insurance suffer[ed] from Robichaux’s fraud.”[227] The court reasoned that the prosecutors remained “within the bounds of reasonableness” because they were simply “impressing upon the jury the seriousness of the charges” which involved “complicated financial transaction[s].”[228] We agree with the Government that if the statements in Robichaux were not prejudicial, then neither are the ones here. The statements are similar and so is the complicated nature of the transactions and fraud. Jacob and Shah argue that the prosecution personally attacked the defendants. Jacob argues that the prosecution “compar[ed] him to a drug dealer.” The prosecution had stated during closing argument that “[m]ost criminals pay their taxes. Drug dealers pay their taxes.” Even if this juxtaposition did constitute an improper remark, Jacob has not shown how it substantially prejudiced him such that reversal is warranted. Shah argues that the Government’s alleged personal attacks launched against Rimlawi were improper and prejudicial. A prosecutor described Rimlawi and his attorney as “cut from the same sleeve. Dirty, nasty.” The Government “regrets” this statement, but it argues that it is not clear Shah has standing to object to a statement made about Rimlawi. Also, even if Shah does have standing, the Government argues that he cannot prove that he received an unfair trial as a result. Rimlawi does not object to the statements made concerning him. Even assuming those remarks were improper and that Shah has standing to object, we agree with the Government that Shah cannot clear the high burden of plain error review and reverse his conviction. Relatedly, even assuming some of the other objected-to statements amounted to misconduct, the defendants have not carried their burden of showing substantial prejudice. The evidence against these defendants was strong, these allegations of misconduct occurred solely during closing argument, and the court offered several limiting instructions throughout the trial. Defendants have not shown that, taken together, the “remarks cast serious doubt on the correctness of the jury’s verdict.”[229] XVI Shah next argues that the district court erred in applying the abuse-of-trust sentencing enhancement to his sentence, but the court did not clearly err. The district court imposed a two-level enhancement under U.S.S.G. § 3B1.3. The enhancement applies “[i]f the defendant abused a position of public or private trust . . . in a manner that significantly facilitated the commission or concealment of the offense.”[230] Shah does not dispute that he occupied a position of trust. His only argument is that he did not use it to facilitate significantly any crime he may have committed. We review for clear error, upholding the enhancement “so long as it is plausible in light of the record as a whole.”[231] We see no clear error in the district court’s finding that Shah used his position of trust to facilitate his crime. He does not dispute that he occupied a position of trust as his patients’ surgeon, and offered Forest Park as a facility where those patients could have their surgeries performed. He was then paid for that referral contrary to law. Shah points to the fact that the sentencing memorandum discusses how Shah was different because he treated DOL patients. Shah argues that the memorandum then ignored that difference by saying he “still took a kickback.” Shah calls the district court’s alleged failure to account for this difference nonsensical because the district court’s omnibus order applied the enhancement to the other surgeons because they were lying to private patients and private insurers. But Shah provides no reason why his enhancement should be any different just because he lied to federal as opposed to private patients. We may affirm “on any basis supported by the record.”[232] The record is clear that Shah used his position as a referring surgeon to facilitate the kickback scheme for which he was convicted. XVII The defendants argue that the district court erred by including proceeds from private-pay surgeries in its calculation of the improper benefit conferred by the kickback scheme. But the district court did not err because the bribes for private insurance patients occurred in the same course of criminal conduct as the bribes for federal-pay patients. The calculation was also otherwise reasonable. At sentencing, the Government requested and the court applied, the sentencing enhancement found at USSG § 2B4.1. That enhancement applies to bribery and kickback cases and enhances the sentence based on the “value of the improper benefit . . . conferred.”[233] That value is measured by  “deducting direct costs from the gross value received.”[234] Direct costs are “all variable costs that can be specifically identified as costs of performing” the bought surgeries.[235] Variable overhead costs generally are not direct costs because they usually “cannot readily be apportioned[,] . . . [and] sentencing courts are not required to make precise calculations.”[236] The difference in cost is also usually de minimis.[237] Indirect (fixed) costs, such as rent and debt obligations, are not deducted from the value of the improper benefit.[238] Henry, Shah, Jacob, and Forrest argue that the court erred in determining the improper benefit amount for purposes of the sentence enhancement found at USSG § 2B4.1. They make two primary arguments: (1) that the district court improperly included the proceeds from Forest Park’s private-insurance patients in its calculation; and (2) that the court calculated the direct-cost reduction incorrectly. Henry and Shah preserved all their arguments below. Forrest did not preserve any, and her claim is reviewed for plain error. Jacob preserved at least some of his argument, but he raises an additional argument on appeal that he did not raise below. That additional argument is reviewed only for plain error. For preserved claims, we review the district court’s interpretation of the guidelines de novo and its factual findings for clear error.[239] There is no clear error if the court’s calculation is plausible; we give district courts wide latitude to calculate the correct amount; and the amount “need only [be] a reasonable estimate . . . based on available information.”[240] We begin with whether the private-pay patient proceeds are properly within the calculation and then turn to whether that calculation was otherwise reasonable. A The improper-benefit sentence enhancement scales according to the amount of the improper benefit received.[241] The greater the improper benefit received, the greater the sentence enhancement. Here, the district court’s calculation of the improper benefit included not only the benefit received from federal-pay surgeries but also from private-pay surgeries. Shah, Forrest, and Jacob contend that the AKS conspiracy involved only federal patients, so the improper-benefit calculation cannot include private-pay patients. Won also attempts to raise this argument but he does so in a single sentence unsupported by caselaw or record citations and has forfeited it.[242] The Government raises two counterarguments. First, it says that the conspiracy was broad enough to encompass private-pay patients. The Government argues that the federal patients served merely to satisfy the jurisdictional hook of the AKS, and that the defendants conspired more broadly to receive remuneration in exchange for referring patients to Forest Park. This conduct, the Government argues, is a conspiracy to violate the AKS because the defendants need not have knowledge of the federal status of their patients, see supra Part II(A). Second, the Government argues that even if the private-pay surgeries were not themselves part of the conspiracy, they were still relevant conduct under the sentencing guidelines and could be factored into the calculation.[243] The Government argues that U.S.S.G. § 1B1.3 requires the court to determine the enhancement based on “all acts and omissions, committed, aided, abetted, counseled, commanded, induced, procured, or willfully caused by the defendant” that either “occurred during the commission of the offense” or “were part of the same course of conduct or common scheme or plan as the offense of conviction.”[244] The Government argues that the private-pay patient kickbacks occurred during the commission of the offense and were part of the same scheme. Shah and Forrest respond that the private-pay patients were not part of the same common scheme because they involved different victims. As in Part II(A), we disagree with the Government’s argument that the federal healthcare program reference in the AKS is only a jurisdictional hook, knowledge of which is not necessary for conviction. The defendants needed to have knowledge that services provided to referred patients may be paid in whole or part by federal healthcare programs. The private-pay surgeries were relevant conduct under U.S.S.G. § 1B1.3 and properly included within the calculation. The sentencing guideline is broad, defining relevant conduct to include “all acts and omissions” that occurred “during the commission of the offense” or as “part of the same course of conduct or common scheme.”[245] “An unadjudicated offense may be part of a ‘common scheme or plan’ if it is ‘substantially connected to the offense of conviction by at least one common factor, such as common victims, common accomplices, common purpose, or similar modus operandi.’”[246] While it may be a close call whether the private-pay surgeries “occurred during the commission of the offense,” they certainly involved the same accomplices (Smith, Burt, and Beauchamp), were completed for the same purpose (bilk insurance providers, whether private or federal, for a high reimbursement rate), and operated with the same modus operandi (pay surgeons to refer surgeries to Forest Park and then use Jacob’s pass-through entity to launder the money).[247] The district court did not err in finding that the private-pay surgeries were part of the same common scheme as the federal-pay surgeries. Shah and Forrest have no answer for this other than an argument that the private-pay surgeries involved different victims, but that does not matter given the substantial overlap of the crimes in all other ways.[248] The defendants also argue that the private-pay surgeries were not relevant conduct because relevant conduct must be criminal, and Jacob argues that the Government never requested a relevant conduct finding in the PSR.[249] Both arguments fail. First, the Government identified several statutes that the private-pay surgeries may have violated. The district court recognized that the Government “ha[d] proven by a preponderance of the evidence” the relevant conduct with which it sought to enhance the sentence. Second, Jacob’s argument is unpreserved, so we review only for plain error, and he is incorrect that the Government did not bring up relevant conduct—it did. So did the PSR. The district court did as well. Finally, Jacob raises a challenge that his enhancements were based on acquitted conduct in violation of the Sixth Amendment right to trial by jury.[250] He argues that sentences that consider acquitted conduct necessarily diminish the jury trial right. In rebuttal, the Government maintains first that Jacob was not acquitted of conspiracy to violate the Travel Act despite being acquitted of the substantive Travel Act counts. It further argues that under this court’s precedent, even acquitted conduct can be the basis of an enhancement so long as the district court finds that the defendant engaged in the conduct by a preponderance of the evidence.[251] While distinguished jurists have questioned the constitutionality of using acquitted conduct for sentencing enhancements,[252] this court has previously recognized that the Supreme Court’s holding in United States v. Watts[253] forecloses Sixth Amendment challenges to the use of acquitted conduct at sentencing.[254] In United States v. Hernandez,[255] we specifically noted that “[Sixth Amendment] challenges are foreclosed under our precedent” and that “the sentencing court is entitled to find by a preponderance of the evidence all the facts relevant to the determination of a sentence below the statutory maximum.”[256] For this reason, Jacob’s argument is unavailing. The record reflects that the district court considered Jacob’s arguments against the use of acquitted conduct, as well as the applicable guidelines range. Jacob thus has not shown that the district court erred when it enhanced his sentence based on acquitted conduct. B Shah, Forrest, and Henry object to the district court’s calculation of the direct-cost reduction. The district court analyzed the hospital’s direct costs as defined by this court’s Landers[257] formula. It looked to costs tied directly to the surgeries performed, i.e., supplies used in the surgery that could not be reused at a later surgery. It determined that the direct costs averaged out to about 21.48% of the total amount Forest Park received in reimbursements. The total amount received in reimbursements, the court reasoned, was the starting place in determining the improper benefit received, and no party challenges this. Shah and Forrest challenge only the 21.48% reduction, arguing that it should be a reduction of 94.2% instead. They arrive at their figure based on the hospital’s net profit margin on the theory that the court had to deduct all costs attributable to the surgery such that the only amount left is the hospital’s net profit. We rejected this exact argument in Landers and do so again.[258] Henry brings a narrower argument, contending that the district court erred because it did not account for the salaries of hospital staff. But again, his argument runs against this court’s holding in Landers that “variable overhead costs that cannot easily be identified” are not direct costs.[259] Although we did not explicitly include staff salaries in the definition of variable overhead costs, they will usually fall within that category of costs. Like rent, debt obligations, and other general overhead costs, staff salaries are not likely to change much because of a specific surgery. Regardless of how many surgeries are performed, those salaries are still paid. In this way, the salaries are costs “incurred independently of output” and not deductible under Landers.[260] Henry has not established that the salaries are not independent of output. Henry’s other arguments to the contrary are unavailing. He cites a study that included salaries as a measure of direct costs, but the study is inapposite. “Direct costs” has a very broad meaning when used in an accounting sense, sufficient even to include staff salaries, but we rejected that definition in Landers.[261] Henry’s citation to United States v. Ricard[262] is similarly inapplicable. There, we reversed because the district court failed to account for any direct costs at all.[263] We never reached the question of whether salaries should be included in direct costs. C Finally, Shah and Forrest briefly argue that the district court erred by “shift[ing] between bribery and fraud theories whenever doing so would increase the sentence.” It is not entirely clear what either defendant is arguing. They do not identify any violation, statutory or constitutional. They do not cite any caselaw. They do not provide record citations. Moreover, the district court only ever applied the bribery guidelines. Any argument that the court misapplied the guidelines has been dealt with above. Any further argument Shah and Forrest may have is forfeited.[264] XVIII Won, Rimlawi, Henry, Jacob, Shah, and Forrest all challenge their restitution amounts. Burt also challenges a part of his restitution judgment. We find no error. Shah, Jacob, Rimlawi, and Won preserved error. Their claims are reviewed de novo as to the legality of the award[265] and method of calculating loss.[266] We review the final restitution amount for abuse of discretion and any factual findings for clear error.[267] We “may affirm in the absence of express findings ‘if the record provides an adequate basis to support the restitution order.’”[268] Forrest did not preserve error, so her claim is reviewed for plain error.[269] Henry and Jacob argue that the Mandatory Victims Restitution Act (MVRA) does not apply to their count-one conviction because it was not “an offense against property,” but they did not preserve this argument. The defendants argue that their claim is reviewed de novo. They base their argument primarily on United States v. Nolen[270] in which a panel of this court reviewed such a claim de novo.[271] United States v. Inman,[272] however, predates Nolen and applied plain error review to such a claim.[273] The Government argues that Inman controls under the rule of orderliness.[274] The Government further argues that Nolen was wrongly decided because it relied on authority that reviewed only for plain error. Relying on the rule of orderliness, we review Henry and Jacob’s unpreserved argument for plain error under Inman.[275] We express no opinion as to whether Nolen was correctly decided, only that it misapplied the rule of orderliness. We turn to the merits of the argument now. A Henry and Jacob argue that the MVRA does not apply to their count-one convictions of conspiracy to violate the AKS because the conspiracy charge did not have fraud or deceit as an element of the crime. They argue that this court should apply the categorical approach to determine whether an offense is an offense against property for purposes of the MVRA. This is a matter of first impression in this circuit, but every other circuit to have addressed this question has determined that the categorical approach does not apply to the MVRA.[276] Neither defendant disputes that, at least as alleged in the indictment, their conduct deprived private insurance companies of property by means of fraud or deceit. But they claim that this is irrelevant because the court must employ the categorical approach and look to the elements of the statute of conviction (18 U.S.C. § 371) to determine whether the MVRA applies. They conclude that no element of conspiracy involves fraud or deceit, so the MVRA does not apply. They further argue that the language of the MVRA mirrors that of other statutes the Supreme Court has held require categorical interpretation. But we find the reasoning of our sister circuits more persuasive on this point. The MVRA provides that restitution must be paid “for[] any offense . . . that is . . . an offense against property under [Title 18] . . . including any offense committed by fraud or deceit . . . in which an identifiable victim or victims has suffered a . . . pecuniary loss.”[277] As the Second Circuit explained, the “committed by fraud or deceit” prong of the MVRA “refers to the way in which some offenses ‘against property’ are ‘committed.’”[278] This “suggests that the way the crime is carried out is relevant to its application.”[279] Further, the statute makes no reference to any elements of a crime against property. This stands in stark contrast to statutes like 18 U.S.C. § 16, which takes an explicit elements-based approach to defining crimes of violence.[280] The categorical approach is inappropriate for this statute and “the [district] court may look to the facts and circumstances of the offense of conviction to determine if the MVRA authorizes a restitution order.”[281] The MVRA is applicable here. The defendants’ “facilitation of . . . payments . . . for phantom work” and general pattern of making and accepting bribes is textbook fraud or deceit.[282] Further, neither defendant objects that at least on its face the indictment alleges that insurance companies suffered pecuniary harm. For further discussion of the private insurers, see below. B Shah, Jacob, Rimlawi, Won, Forrest, and Henry argue that private insurers were not proper victims under the MVRA and that their restitution amounts must be reduced accordingly. Under the MVRA, “victim” means: a person directly and proximately harmed as a result of the commission of an offense for which restitution may be ordered including, in the case of an offense that involves as an element a scheme, conspiracy, or pattern of criminal activity, any person directly harmed by the defendant’s criminal conduct in the course of the scheme, conspiracy, or pattern.[283] The district court found that the private insurers were victims under the Act because they paid inflated claims to Forest Park as a result of the defendants’ surgery-buying scheme. The defendants do not dispute that the private insurers suffered direct and proximate harm. Their only argument, mirroring that found in Part XVII, is that the private insurers were outside the conspiracy’s scope. For the same reasons as outlined above in Part XVII, the private insurers were within the scope of the conspiracy. While true that it was the presence of federal insureds that granted federal jurisdiction in this case and was necessary for conviction, the conspiracy was one to steer patients to Forest Park by way of buying surgeries. It covered both private and federal patients. In fact, as the defendants themselves argue, they were expressly trying to avoid federal patients. They targeted private patients directly. Further, the MVRA’s definition of “victim” is quite broad such that even assuming the private-pay patients were not part of the conspiracy, we would still affirm. As above, the MVRA defines victims as those harmed “in the course of the . . . conspiracy.”[284] The private insurers were harmed at the same time and in the same manner as the federal insurers because the bribe payment that was the basis for the inflated claims was the same no matter whether the patient was insured federally or privately. This overlap, similar to the analysis in Part XVII, brings the private insurers into the role of victim.[285] We have held, in United States v. Gutierrez-Avascal,[286] that the driver of a car hit by a fleeing member of a marijuana conspiracy was a victim of the marijuana conspiracy.[287] There is very little daylight between the rationale there and here. As the defendants conspired to buy surgeries, private insurers suffered direct losses just as the driver in Gutierrez-Avascal did. The defendants’ arguments to the contrary are unavailing. They largely reiterate their arguments that private patients and insurers were not part of the conspiracy. We have already rejected this argument. They also argue that the various Travel Act acquittals somehow bring the private insurers out of the role of victim, but for the reasons explained above, the private insurers are victims of the count one, AKS conspiracy, so the Travel Act acquittals mean nothing in this context. C Only Rimlawi challenges the final amount of restitution ordered against him. His main argument is that the district court did not properly address his restitution arguments. He did not raise this argument below when the district court at sentencing asked if there were “any unaddressed issues.” Accordingly, we review it for plain error.[288] The PSR and the Government put forward a detailed explanation as to the restitution amount for each defendant. The record has “an adequate basis” for the restitution amount.[289] We may affirm on that basis.[290] Further, the district court is granted “wide latitude” in calculating the final amount which need only be “a reasonable estimate.”[291] Rimlawi has done nothing to show how a different treatment of his restitution arguments would result in a different amount, nor how a different amount would substantially affect his rights. D Finally, seizing upon a recent dissent from a denial of certiorari, Rimlawi, Shah, Henry, and Forrest argue that a jury must find the restitution amount beyond a reasonable doubt. They concede that this issue is foreclosed—they seek only to preserve it for further review.[292] We will not address it further. XIX Finally, Won and Rimlawi argue that the district court erred in calculating the forfeiture amount. We find no error. We review the legality of forfeiture de novo.[293] The criminal forfeiture statute, 18 U.S.C. § 982, requires the court, “in imposing sentence on a person convicted of a Federal health care offense, . . . [to] order the person to forfeit property, real or personal, that constitutes or is derived, directly or indirectly, from gross proceeds traceable to the commission of the offense.”[294] The analytical inquiry is whether the defendant would have received the property “but for” his criminal conduct.[295] The basis of Won and Rimlawi’s argument is essentially the same as their argument as to restitution. They claim that the private insurers were not part of the conspiracy and therefore any proceeds derived therefrom do not fall within the forfeiture statute. As explained above, receiving kickbacks for the privately insured patients was part of the conspiracy. Won and Rimlawi would not have received their bribe money “but for” their referrals to Forest Park.[296] These referrals included not only private but also federal patients. The agreement, however, was the same for both sets of patients—the surgeons referred patients and the hospital paid them per patient. But for that illegal conduct of conspiring to send the patients to Forest Park under a handshake deal for a kickback, the surgeons would not have received their proceeds. As above, the bribe money did not differentiate between federal patients or private patients—the agreement and reimbursement were the same for both. The surgeons’ conduct falls squarely within the realm of forfeiture.[297] Won and Rimlawi’s arguments to the contrary are unavailing. Largely, they repeat arguments already dealt with above. They hang their hat on the Travel Act acquittals, but again, any acquittal there is meaningless here because the private insurers were part of the count-one AKS conspiracy conviction. Thus, forfeiture of proceeds derived from their loss is still “tied to the specific criminal acts of which the defendant was convicted.”[298] * * * For the foregoing reasons, we AFFIRM.