United States v. Ihab Steve Barsoum

Ihab “Steve” Barsoum appeals his convictions and 204-month sentence, imposed after a jury convicted him of one count of conspiring to dispense Oxycodone not for a legitimate medical purpose and not in the usual course of professional practice, in violation of 21 U.S.C. §§ 841(a)(1), (b)(1)(C), and 846, and five counts of distributing Oxycodone outside the course of professional practice, in violation of 21 U.S.C. §§ 841(a)(1) and (b)(1)(C). Bars-oum appeals several of the District Court’s rulings before, during, and after trial, including its drug-quantity findings, which it *1326 based on estimates derived from the record evidence. For the reasons stated below, we affirm.

I. FACTS AND PROCEDURAL HISTORY

This case centers around a complex web of schemes Barsoum implemented with his coconspirators to illegally distribute tens of thousands of Oxycodone pills from several pharmacies between sometime in 2007 and July 2011. After Barsoum’s trial and sentencing hearing, the District Court made drug-quantity findings for sentencing and ultimately found Barsoum responsible for 56,000 30-mg Oxycodone pills. The overall drug-quantity finding was based on the number of pills across four categories: (1) sales to an associate named Scott-16,000 pills; (2) fraudulent prescriptions in the name of one Dr. Belsole (the “Belsole prescriptions”) — 11,000 pills; (3) sales to an associate named Stevens — 24,-000 pills; and (4) sales to Stevens during controlled buys orchestrated by the Drug Enforcement Administration (“DEA”)— 5,000 pills.

A. Barsoum’s First Pill Scheme (St. George’s Pharmacy)

In 2006, Barsoum began working at St. George’s Pharmacy, which was run by Barsoum’s cousin, Wahba, in Pinellas County, Florida. Wahba and a few drug dealers had already established a pill racket at St. George’s by the time Barsoum started there. Barsoum’s soon-to-be drug associate, Pat Stevens, and others would bring prescriptions under fake patient names to be filled by Wahba. Wahba would then issue large amounts of Oxyco-done and OxyContin without asking for identification. Soon after starting at St. George’s, Barsoum also began accepting and filling the falsified prescriptions. Bars-oum would fill prescriptions for Stevens as often as once a day. He never asked for identification even though Stevens’ prescriptions used various patient names. In some instances, Stevens would hand Bars-oum a pile of prescriptions all at once, paying large amounts of cash at a set rate per pill. The pills Barsoum distributed at St. George’s Pharmacy were not figured into his Sentencing Guidelines calculation and are not at issue in this appeal. Nevertheless, St. George’s is where Barsoum got his start — where he learned the tactics he would employ over the next several years at other pharmacies to conceal his illegal pill distribution schemes.

B. Barsoum’s Pill Scheme with Scott (Trinity Pharmacy)

In 2007, Barsoum opened his own pharmacy, Trinity Pharmacy, where he soon began accepting falsified prescriptions from an Oxycodone addict and dealer named Christopher Scott. Barsoum coordinated the scheme closely with Scott to avoid DEA suspicion. Barsoum provided Scott actual doctors’ names and DEA numbers 1 so that the prescriptions appeared genuine. Whenever Scott presented Bars-oum with an incorrect DEA number or a prescription written for suspiciously too many pills, Barsoum would tell Scott to rewrite the prescription. Scott would then leave Trinity Pharmacy, correct the prescription, and return minutes later to fill it. Barsoum educated Scott on several other strategies for avoiding DEA scrutiny as well.

Despite Barsoum’s best efforts, the pill scheme drew the attention of the authorities. The local police department began *1327 investigating Scott, whom they suspected of dealing Oxycodone. The police approached Barsoum for information about Scott and asked that he notify them when he saw Scott again. Instead, Barsoum alerted Scott to the investigation and never notified the police.

Scott was eventually, and inevitably, arrested in 2008. At trial, Scott testified that "Barsoum filled 4 prescriptions at a time for him, usually for 240 pills each, about once a month, 2 for 17 months (4 x 240 x 1 x 17 = 16,320). The District Court attributed 16,000 Oxycodone pills to Barsoum for this scheme. These pills comprise the first category of the District Court’s drug-quantity findings for Bars-oum’s Guidelines calculation. We further discuss the finding in this category below. See infra Part II.D.ii.a.

Our takeaway from Barsoum’s scheme with Scott is that through teaching Scott how to avoid the DEA, and through dispensing such large quantities of pills, Barsoum had positioned himself as the centerpiece of a sophisticated Oxycodone distribution conspiracy.

C. Barsoum’s Pill Scheme with Stevens (Trinity Pharmacy)

In 2008, Barsoum’s prior associate from the St. George’s scheme, Stevens, also began filling falsified prescriptions at Trinity Pharmacy. Barsoum coordinated with Stevens, as he did with Scott, to embellish prescriptions so that they would appear genuine and elude DEA suspicion. Bars-oum provided Stevens the names, addresses, phone numbers, and DEA numbers of actual doctors. Barsoum and Stevens communicated via disposable, non-registered phones to further cover their tracks. Between 2009 and 2010, Stevens visited Trinity Pharmacy up to three times a week. He testified at trial that he obtained between 300 and 700 pills each week. He paid Barsoum $4 per pill and resold them for $10 each.

The DEA arrested Stevens in November 2010. The DEA confiscated approximately $203,000 in cash, which Stevens later stated was mostly drug money from reselling Barsoum’s pills. He also gave three different post-arrest statements as to the number of pills he had obtained from Barsoum. Stevens gave widely varying estimates of 700 to 800 pills per week for the year, only 500 pills per month for the year, and only 8,000 pills total. The pills Barsoum illegally passed Stevens before Stevens began cooperating with the DEA were figured into his Guidelines calculation as the third category of pills. We further discuss the District Court’s finding in this category below. See infra Part II.D.ii.c.

D. DEA Controlled Buys throuyh Stevens (Platinum Pharmacy)

Barsoum later , opened another pharmacy, Platinum Pharmacy, where he continued distributing to Stevens. Barsoum was unaware that Stevens had begun cooperating with the DEA in a series of undercover, controlled buys. Each undercover buy operated the same way as Barsoum’s prior schemes. Stevens exchanged falsified yet authentic-looking prescriptions and large amounts of cash for large quantities of Oxycodone. Just as in his previous schemes, Barsoum accepted and filled the prescriptions without ever asking for identification. He also created a phone record of feigned prescription verification calls. He pretended to call doctors’ offices to verify the prescriptions, but he was actually calling his own throwaway phones. The pills Barsoum distributed during the controlled buys were figured into his Guidelines calculation as the fourth category of *1328 pills. 3 Barsoum does not contest this category on appeal.

E. Belsole Prescriptions

The DEA arrested Barsoum in October 2011, searched Platinum Pharmacy, and seized over sixty prescriptions dated February 2011 to July 2011. The prescriptions were purportedly signed by one Dr. Belsole and accounted for over 11,000 Oxy-codone pills. Evidence suggested that the Belsole prescriptions were fraudulent and were used similarly to the fraudulent prescriptions in Barsoum’s other schemes. Thus after hearing the evidence, the District Court attributed the 11,000 pills authorized by the Belsole prescriptions to Barsoum as the second category of pill amounts for sentencing. We discuss these further below. See infra Part II.D.ii.b.

Ultimately, the jury convicted Barsoum of the conspiracy count and all five distribution counts charged in the indictment. The District Court calculated Barsoum’s Guidelines range at 235 to 293 months after aggregating the four categories of drug-quantity findings. The District Court departed downward and sentenced Barsoum to 204 months.

II. DISCUSSION

A. Denial of the Motion to Suppress and Request for Franks Hearing

Below, Barsoum filed a motion 4 to suppress evidence and a request for a hearing pursuant to Franks v. Delaware, 438 U.S. 154, 98 S.Ct. 2674, 57 L.Ed.2d 667 (1978), which the District Court denied. Barsoum alleged that the special agent who swore out the affidavit either recklessly or intentionally omitted information from the warrant application. Barsoum contends on appeal that the District Court erred in denying his request for a Franks hearing, because “[h]ad there been a Franks hearing, the defense would have been able to show that the affidavit deliberately omitted material information as to make its probable cause presentation meaningless.” Appellant’s Br. at 31. As to his motion to suppress, Barsoum points on appeal to a flurry of supposed deficiencies in the warrant application and affidavit. He believes these deficiencies negated probable cause to search Platinum Pharmacy for any evidence.

i. Standard of Review

We generally review a district court’s denial of an evidentiary hearing for an abuse of discretion. United States v. Arbolaez, 450 F.3d 1283, 1293 (11th Cir.2006). We have not, however, explicitly adopted a standard of review for a district court’s denial of a Franks hearing. Id. Nonetheless, abuse of discretion review is appropriate. See United States v. Sarras, 575 F.3d 1191, 1218 (11th Cir.2009).

A district court’s denial of a motion to suppress is a mixed question of law and fact. United States v. Delaney, 502 F.3d 1297, 1304 (11th Cir.2007). We review a district court’s factual findings for clear error and review its application of law to the facts de novo. Id.

ii. Analysis

A Franks hearing is warranted where a defendant “makes a substantial preliminary showing” that an affiant made intentionally false or recklessly misleading statements (or omissions), and those statements are “necessary to the finding of probable cause.” Franks, 438 U.S. at 155-56, 98 S.Ct. 2674. ‘When assessing *1329 whether the alleged false statements and omissions were material, the trial court is to disregard those portions of the affidavit which the defendant has shown are arguably false and misleading.” Sarras, 575 F.3d at 1218 (citations omitted). “The defendant bears the burden of showing that, absent those misrepresentations or omissions, probable cause would ' have been lacking.” Id. (citations and internal quotation marks omitted).

We need not address the statements Special Agent Zdrojewski allegedly omitted from the affidavit. The heart of Barsoum’s argument is that a Franks hearing would have allowed him to show that the information was deliberately omitted. This argument misunderstands the Franks standard. Franks requires that a defendant make a substantial 'preliminary showing that statements or omissions are intentionally false or recklessly misleading, and that those statements or omissions altered the probable cause showing. 438 U.S. at 171, 98 S.Ct. 2674. The allegations of deliberate omission “must be more than conclusory and must be supported by more than a mere desire to cross-examine.” Id. To the extent Barsoum argues that only a Franks hearing could have shown that information was deliberately omitted from the Affidavit, he concedes that he failed to make a substantial preliminary showing of deliberate falsity and omission. See id. The District Court fully addressed Bars-oum’s allegations of falsity and concluded that they were unpersuasive. For these reasons, we find no abuse of discretion in the District Court’s denial of Barsoum’s request for a Franks hearing.

As to the flurry of supposed deficiencies in the warrant application and affidavit, after a de novo review of the record, the motion to suppress, Barsoum’s appeal brief, and the Magistrate Judge’s Report and Recommendation, 5 we also find Bars-oum’s arguments unpersuasive. Bars-oum’s arguments raise no substantial issue for consideration. 6 Because our analysis of Barsoum’s motion to suppress tracks the Magistrate Judge’s extensive analysis, it is unnecessary to address the many supposed deficiencies Barsoum points to. We conclude that the application and affidavit supported a finding of probable cause. The District Court correctly denied Bars-oum’s motion to suppress.

B. Denial of the Motion for Judgment of Acquittal

Barsoum argues the District Court erred in denying his motion for judgment of acquittal because the Superseding Indictment 7 charged one conspiracy to distribute Oxycodone between an unknown date in'2007 and July 2011, whereas he believes the government proved at least four separate conspiracies at trial. He argues the evidence showed that (1) Scott, (2) Stevens before his arrest, (3) Stevens after he began cooperating with the DEA, and (4) “the unknown [person] responsible for the Belsole prescriptions,” Appellant’s Br. at 57, all had separate conspiracies and that he was improperly convicted of a single overarching conspiracy.

i. Standard of Review

We review the denial of a motion for judgment of acquittal de novo, but viewing the' evidence “in the light most favorable to the government, with all rea *1330 sonable inferences and credibility choices made in the government’s favor.” United States v. Ortiz, 318 F.3d 1030, 1036 (11th Cir.2003) (citations omitted).

ii. Analysis

We will not reverse a conviction because a single conspiracy is charged in the indictment yet multiple conspiracies were proved at trial unless the variance is (1) material and (2) substantially prejudiced the defendant. United States v. Richardson, 532 F.3d 1279, 1284 (11th Cir.2008). Because the jury determines the factual question of whether the evidence established a single conspiracy, a “material variance” between the indictment and the evidence results only where there is no evidentiary foundation for the jury to find a single conspiracy. Id. To assess whether a material variance resulted, we must consider: “(1) whether a common goal existed; (2) the nature of the underlying scheme; and (3) the overlap of participants.” Id. To demonstrate “substantial prejudice,” a defendant must show either:

1) that the proof at trial differed so greatly from the charges that [he] was unfairly surprised and was unable to prepare an adequate defense; or 2) that there are so many defendants and separate conspiracies before the jury that there is a substantial likelihood that the jury transferred proof of one conspiracy to a defendant involved in another.

Id. at 1286-87 (quoting United States v. Calderon, 127 F.3d 1314, 1327 (11th Cir.1997)).

a. Material Variance

A hub-and-spoke conspiracy is one in which a “ ‘key man’ [or hub] directs and coordinates the activities and individual efforts of various combinations of people.” Id. at 1284-85; United States v. Edouard, 485 F.3d 1324, 1347 (11th Cir.2007). Bars-oum’s rationale is that he, was the spoke of several different hub-and-spoke conspiracies, see Appellant’s Br. at 35, while the individuals he passed thousands of pills to and taught how to avoid DEA scrutiny were the hubs of those conspiracies. He advances this rationale because in cases in which the defendant was a spoke, rather than a key man or hub, he must have knowledge of the other spokes in order to be properly convicted of a single, overarching conspiracy with those other spokes. See, e.g., Kotteakos v. United States, 328 U.S. 750, 754-55, 66 S.Ct. 1239, 1242-43, 90 L.Ed. 1557 (1946); United States v. Chandler, 388 F.3d 796, 807 (11th Cir.2004). However, where the defendant was the key man or hub, a jury may properly conclude that a single conspiracy was proved at his trial, even where the spokes had no knowledge of each other. See Richardson, 532 F.3d at 1284-85; Edouard, 485 F.3d at 1347. “It is irrelevant that particular conspirators may not have known other conspirators,” because when a “ ‘key man’ directs and coordinates the activities and individual efforts of various combinations of people,” the hub may be properly convicted of a single conspiracy. Richardson, 532 F.3d at 1284-85, 1286.

Thus the question is whether, when viewing the evidence in the light most favorable to the government, the jury had an evidentiary basis to conclude that Barsoum was the key man or hub of a single conspiracy.

We first consider whether a common goal existed. “[C]ommon for the purposes of this test means similar or substantially the same.... ” Richardson, 532 F.3d at 1285 (internal quotation marks omitted). The jury heard evidence regarding Bars-oum’s scheme with Scott, that: (1) Scott fraudulently made prescriptions under a real doctor’s name — Dr. Jesus Mar *1331 tinez — using the doctor’s- DEA number; (2) Scott filled prescriptions at -Barsoum’s Trinity Pharmacy under fictitious or stolen names and without presenting identification; (3) Barsoum never turned Scott away even though he presented no identification; (4) Scott filled prescriptions under those varying names approximately once every twenty days; (5) Barsoum usually passed 240 30-mg Oxycodone pills to Scott each visit (the equivalent of one pill every two hours for twenty days); (6) Barsoum coached Scott with tactics to avoid the DEA’s suspicion, such as reducing extraordinarily large quantities of pills, including auxiliary drugs an actual physician might prescribe in conjunction with Oxycodone, and correcting wrong DEA numbers; and (7) the last time Scott went to Trinity Pharmacy, Barsoum warned him that the police were looking for him.

As for Barsoum’s scheme with Stevens, the jury heard evidence that: (1) Stevens had been a part of a pill ring that filled prescriptions under various names for Oxycontin at St. George’s, where Barsoum was previously a clerk; (2) Barsoum notified Stevens in 2008 that he was opening his own pharmacy and was willing to continue a pill ring with him; (3) Barsoum coached Stevens with information including a doctor’s name, DEA number, address, and phone number; (4) Barsoum gave Stevens blank prescriptions; (5) Barsoum and Stevens communicated through non-registered disposable phones; (6) Stevens visited Barsoum’s pharmacy up to three times a week between October 2009 and November 2010, obtaining 300 to 700 pills each week; and (7) Stevens eventually cooperated with the DEA to carry out controlled buys from Barsoum wherein Stevens used doctors’ names and DEA numbers that he and Barsoum had shared with each other.

Viewing the foregoing in the light most favorable to the government, we conclude that a common goal existed between Bars-oum and each of his coconspirators. Specifically, Barsoum shared with each of his coconspirators the common goal of filling unauthorized prescriptions for large amounts of Oxycodone and Oxycontin at his pharmacies, through the use of fictitious names, stolen doctors’ information, and otherwise fraudulent prescriptions.

Second, we consider the nature of the underlying - scheme. Viewing Barsoum’s aforementioned schemes in the light most favorable to the government, we conclude that each sub-scheme was part and parcel of Barsoum’s overarching conspiracy. Specifically, each of Barsoum’s coconspira-tors worked with him to use fictitious names, stolen doctors’ information, and fraudulent prescriptions to disguise his unauthorized distribution of Oxycodone.

Third, Appellant concedes there is an overlap of participants as he overlapped each of the schemes. Appellant’s Br. at 38. Nevertheless we review the evidence de novo in the light most favorable to the government. While there were various players between 2007 and 2011, most notably Scott and'Stévens, Barsoum was the common denominator in each of the schemes. Barsoum played the same role in each scheme — he was the gatekeeper and the hub. The thousands of pills all flowed from Barsoum, who directed and coached his coconspirators to help them all avoid the DEA. Viewed in the light most favorable to the government, the fact that Barsoum overlapped the schemes in this way weighs towards finding a single conspiracy.

Based on our foregoing analysis, we conclude that the jury had an evidentiary basis to find that Barsoum was the key man or hub of a single conspiracy to ille- *1332 gaily distribute Oxycodone between 2007 and 2011. No material variance occurred.

b. Substantial Prejudice

We need not address this issue because the lack of a material variance, alone, is sufficient to affirm the denial of Barsoum’s motion for judgment of acquittal. Nevertheless, we find no substantial prejudice.

Barsoum only argues the confusion of proof theory. 8 Barsoum was the key man of the conspiracy and was the only defendant in his trial. We have held that a defendant cannot be substantially prejudiced under this test if “[he] was the hub of the conspiracy and was tried alone.” Richardson, 532 F.3d at 1287. There is therefore no “likelihood that the jury transferred proof of one conspiracy to a defendant involved in another.” Id.

C. Denial of the Motion to Dismiss the Conspiracy Charge as Duplicitous

Barsoum argues the District Court erred in denying his motion to dismiss the conspiracy charge as duplicitous, for the same reasons he argued there was a material variance. See Appellant’s Br. at 43-45. He believes the evidence presented at trial forced him to defend against several separate conspiracies. Because we have concluded that the jury properly found a single conspiracy was proved at trial, we reject Barsoum’s duplicity arguments.

D. The District Court’s Drug-Quantity Findings Based on Estimates

As we stated before, the District Court found Barsoum responsible for 56,000 30-mg Oxycodone pills, across four categories: (1) the scheme with Scott — 16,000 pills; (2) the Belsole prescriptions — 11,000 pills; (3) the scheme with Stevens — 24,000 pills calculated by estimating 500 pills per week for one year; 9 and (4) controlled buys by Stevens while he was cooperating with the DEA — 5,000 pills.

As to the first two categories, Barsoum argues that he did not enter into an agreement with Scott or the “unknown” individual responsible for the Belsole prescriptions, and that the government failed to prove an overall conspiracy. He therefore argues the District Court clearly erred by including Categories 1 and 2 in his Sentencing Guidelines calculation.

As to the third category, Barsoum argues generally that because the District Court’s finding was based on varying estimates, it was not supported by a preponderance of the evidence. Specifically, he argues the evidence did not support an estimate of 500 pills per week, (1) because Stevens gave shifting estimates of the number of pills he acquired from Barsoum and (2) because Stevens’ testimony and Special Agent Zdrojewski’s testimony conflicted in a few key aspects. Special Agent Zdrojewski is the DEA agent who swore out the affidavit and was integral in investigating and arresting Barsoum. Special Agent Zdrojewski testified at the sentencing hearing about the number of pills attributable to Barsoum. A substantial portion of his testimony was devoted to explaining why Stevens’ estimates differed from his and from each other, and why 24,000 pills was an appropriate estimate.

*1333 Barsoum does not contest the District Court’s finding of 5,000 pills in the fourth category.

i. Standard• of Review

We review a district court’s application of the facts to the Sentencing Guidelines de novo. United States v. Norris, 452 F.3d 1275, 1280 (11th Cir.2006). However, we review drug-quantity, findings only for clear error. United States v. Almedina, 686 F.3d 1312, 1315 (11th Cir.2012), ce rt. denied, — U.S. -, 133 S.Ct. 629, 184 L.Ed.2d 408 (2012); see United States v. Jordi, 418 F.3d 1212, 1214 (11th Cir.2005). “For a finding to be clearly erroneous, [we] must be left with a definite and firm conviction that a mistake has been committed.” Almedina, 686 F.3d at 1315 (quoting United States v. Rothenberg, 610 F.3d 621, 624 (11th Cir.2010)). Arguments not raised on appeal are deemed abandoned. United States v. Cunningham, 161 F.3d 1343, 1344 (11th Cir.1998).

ii. Analysis

When a defendant challenges one of the factual bases of his sentence, such as a drug quantity set forth in the presentence report (“PSR”), the government has the burden of establishing that fact by a preponderance of the evidence. United States v. Rodriguez, 398 F.3d 1291, 1296 (11th Cir.2005). When calculating drug quantities, if the amount seized does not reflect the entirety of the offense, the sentencing court must find the total drug quantity by estimating. United States v. Frazier, 89 F.3d 1501, 1506 (11th Cir.1996). In estimating drug quantities, the sentencing court must consider all acts committed, induced, or willfully caused by the defendant. See U.S.S.G. § lB1.3(a)(l)(A). The sentencing court may base its calculation on evidence showing the average frequency and amount of a defendant’s drug sales over a given time period. Frazier, 89 F.3d at 1506. The calculation may be based on “fair, accurate, and conservative estimates,” but not on mere speculation. Almedina, 686 F.3d at 1316.

‘We accord great deference to the district court’s credibility determinations” of drug-quantity witnesses. United States v. Gregg, 179 F.3d 1312, 1316 (11th Cir.1999). Where evidence gives rise to two reasonable and different constructions, as conflicting witnesses’ testimony may, the fact finder’s choice between the two constructions cannot be clearly erroneous. Almedina, 686 F.3d at 1315.

a. First Category of Pills (Scheme with Scott)

The District Court found Barsoum responsible for 16,000 Oxycodone pills from his scheme with Scott. Barsoum only argues that the District Court erred in attributing the pills to him because he had no agreement with Scott, i.e., that Barsoum and Scott were not coconspira-tors. We reject this argument as we have already concluded that the jury properly found Barsoum guilty of a single conspiracy that included Scott as a coconspirator. Nevertheless, we review the finding in this category for clear error. See Almedina, 686 F.3d at 1315.

We reiterate our discussion of Bars-oum’s scheme with Scott. See supra, Parts I.B, II.B,ii.a. We note specifically that, at trial, Scott testified that he filled 4 prescriptions at a time with Barsoum, usually for 240 pills each, about once a month, for 17 months (4 x 240 x 1 x 17 = 16,320). Barsoum never turned Scott away and never required him to present identification. The District Court heard evidence that Barsoum coached him to include auxiliary drugs on the prescriptions *1334 that a real doctor might prescribe alongside Oxycodone. Scott also testified that Barsoum warned Scott that the police were investigating him. Additionally, the District Court considered the PSR, which stated that “Barsoum supplied Scott with approximately 16,080 dosage units of 30 milligram Oxycodone pills.” PSR Addendum ¶ 14. Under these facts, the District Court did not clearly err in finding by a preponderance of the evidence that Bars-oum was responsible for illegally passing 16,000 pills to his coconspirator, Scott.

b. Second Category of Pills (the Belsole Prescriptions)

Barsoum argues that the District Court erred in including the pills associated with the Belsole prescriptions in his Guidelines calculation, because he had no agreement with “the unknown [person] responsible for the Belsole prescriptions.” Appellant’s Br. at 57. We review the finding of 11,000 pills in this category for clear error. Almedina, 686 F.3d at 1315.

The record included the following evidence on the Belsole prescriptions: (1) the DEA searched Platinum Pharmacy after Barsoum’s arrest and discovered over sixty prescriptions for Oxycodone in the name of Dr. Belsole, totaling over 11,000 30-mg pills; (2) the prescriptions were dated between February 2011 and July 2011 — during the time of the conspiracy for which he was convicted; (3) the prescriptions had been cut unevenly and to awkward sizes, similar to the way Bars-oum cut prescriptions when asked to as part of the investigation; (4) some of the prescriptions included the same fictitious patient names that Stevens had previously presented him; (5)- Barsoum signed the back of some of the prescriptions in the same way that he signed other prescriptions to indicate he had made a verification call to a doctor’s office; (6) the real Dr. Belsole testified that he did not write prescriptions for Oxycodone as part of his practice, that he was unfamiliar with the healthcare facility from which the prescriptions purportedly originated, and that the “Belsole” signature found on the prescriptions was not Ms.

Under these facts, it was not clear error for the District Court to include the Bel-sole prescriptions as relevant conduct of the conspiracy and substantive counts for which he was convicted. See Norris, 452 F.3d at 1280; U.S.S.G. §§ lB1.3(a)(l)-(2). The evidence showed that Barsoum, rather than some unknown person, fraudulently created the Belsole prescriptions and illegally dispensed pills pursuant to them. The District Court stated “[t]here’s no question that those Belsole prescriptions were dispensed[,] ... that [they] were fake[,] ... [or] that [Barsoum] absolutely knew that they were fake.” We agree. The District Court therefore properly included 11,000 Oxycodone pills from the Belsole prescriptions in Barsoum’s Guidelines calculation.

c. Third Category of Pills (Scheme with Stevens)

The District Court attributed 24,000 Ox-ycodone pills to Barsoum for his scheme with Stevens. To make its determination, the District Court considered the evidence at trial, including Stevens’ testimony, and Special Agent Zdrojewski’s testimony at the sentencing hearing. Stevens’ estimates of the number of pills he acquired from Barsoum were internally inconsistent and differed from Special Agent Zdrojew-ski’s testimony in a few aspects. The District Court eventually settled at 24,000 pills — an amount far fewer than the 60,000 pills the PSR recommended under this category.

As we stated before, the DEA arrested Stevens in November 2009. Throughout *1335 the course of several post-arrest statements and trial, Stevens gave four different estimates. First, the DEA took statements from Stevens immediately after his arrest. He told the DEA that he had been obtaining approximately 700-800 pills per week (2,800-3,200 pills per month) for a year and reselling them for $10 each. He also told the DEA that the $203,000 confiscated was mostly drug money from reselling Barsoum’s pills. Second, sometime later, Special Agent Zdrojewski debriefed and interviewed Stevens. He told Stevens it was his practice to ask informants for conservative estimates of drug quantities. He therefore asked Stevens for a “low ball” estimate, to which Stevens replied he had been obtaining only 500 pills per month. Stevens later gave a third estimate in a typed a letter stating he had only obtained 8,000 pills total. He then eventually gave a fourth estimate when testifying at trial — 300-700 pills per week.

Special Agent Zdrojewski gave extensive direct and cross-examination testimony regarding the drug amounts attributable to Barsoum. He corroborated that Stevens originally gave DEA agents an estimate of 700-800 pills per week. He also stated that he specifically asked for a “low ball” estimate when Stevens estimated 500 pills per month. Special Agent Zdrojewski also testified regarding the $203,000, that:

Once looking into it, the amount of money, $10 per pill, you figure ... even if we went off the low end, went off his original debrief when he had said [700] to 800, even if that’s not consistent, if you went just 500 pills per week time four weeks per month, you’re looking at [2,000] — then over the year, that’s 24,000 pills times $10, that’s $240,000.

He had [$]203,000 in the safe, or in the house, and figure expenses and whatever else he’s purchased, it would roughly come back to the original debrief post arrest....

Sentencing Hr’g Tr. 10 at 41. In other words, conservatively estimating 500 pills per week (2,000 pills per month or 24,000 pills for the year) from Stevens’ original 700-800 pills per week estimate, the $203,000 confiscated was consistent with Stevens having sold 24,000 pills at $10 each. The District Court found this evidence “pretty compelling,” and eventually settled at a finding of 500 pills per week over a year, which aggregates to the 24,-000 pills in this category.

Barsoum argues that the District Court’s finding is not supported by a preponderance of the evidence because Stevens’ testimony is internally inconsistent, and because Stevens and Special Agent Zdrojewski gave differing accounts. Bars-oum points to the four, very different estimates: (1) 700-800 pills per week (2,800-3,200 pills per month); (2) 500 pills per month; (3) 8,000 pills total (about 665 pills per month); and (4) 300-700 per week (1,200-2,800 pills per month). He also notes that Stevens had additional income from his business, a gentleman’s club, which would account for some of the $203,000 confiscated.

We find no clear error in the District Court’s determination. We conclude that its estimate was based on a “fair, accurate, and conservative estimate,” Almedina, 686 F.3d at 1316, and supported by a preponderance of the evidence. First, because the government was unable to seize and present as evidence all the pills Barsoum passed to Stevens, the District Court necessarily estimated to make its drug-quantity finding. Frazier, 89 F.3d at 1506. The District Court properly relied on evidence at trial and the sentencing hearing showing the average frequency and number of *1336 pills Barsoum passed Stevens over a one-year period. Id.

We recognize that the average estimates Stevens gave vary widely from each other and from Special Agent Zdrojewski’s 500 per-week estimate. However, the District Court’s finding of 500 pills per week is logically consistent with the 800-700 pills per week Stevens testified to at trial. The finding was also based on the 500 pills per week that Special Agent Zdrojewski suggested as a conservative estimate of Stevens’ original 700-800 per week estimate. As for Stevens’ most discrepant estimate of only 500 pills per month, the District Court heard evidence that he only gave that number when specifically asked for a “low ball” estimate. As for the estimate of 8,000 pills total, Stevens responded at trial that the number seemed lów and he suggested it was likely a mistake. Lastly, despite what Barsoum argues, 500 pills per weék sold for a total $240,000 is consistent with the $203,000 the DEA confiscated, which Stevens stated was derived mostly from reselling Barsoum’s pills.

Given the foregoing, the District Court’s drug-quantity finding of 24,000 pills, based on an estimated 500 pills per week, was supported by a preponderance of the evidence. “We accord great deference to the [District [Cjourt’s credibility determinations.” Gregg, 179 F.3d at 1316. We therefore find no clear error. The District Court was presented with two reasonable yet differing accounts, and based its drug-quantity finding on Special Agent Zdro-jewski’s account, which it found more credible. Almedina, 686 F.3d at 1315. The District Court properly attributed 24,000 Oxycodone pills under this category for Barsoum’s Guidelines calculation.

d. Fourth Category of Pills (DEA Controlled Buys)

The District Court also included in Bars-oum’s Guidelines calculation 5,000 Oxyco-done phis he passed Stevens as part of the DEA controlled buys. Barsoum did not contest these pills at sentencing. Likewise, he advances no argument before us against these pills being included. We therefore deem any argument against these pills abandoned. Cunningham, 161 F.3d at 1344.

For the foregoing reasons, we affirm the overall finding of 56,000 Oxycodone pills attributable to Barsoum for purposes of sentencing.

E. Joinder of the Substantive Counts with the Conspiracy Count

Barsoum argues that joinder of his conspiracy count with his substantive counts in the Superseding Indictment was improper, and that the District Court abused its discretion in denying his motion to sever the counts.

i. Standard of Review

“We undertake a two-step analysis to determine whether separate charges were properly tried at the same time.” United States v. Slaughter, 708 F.3d 1208, 1213 (11th Cir.2013), cert. denied, — U.S. -, 133 S.Ct. 2868, 186 L.Ed.2d 920 (2013) (quoting United States v. Hersh, 297 F.3d 1233, 1241 (11th Cir.2002)). We first review de novo whether counts were properly joined in one indictment under Federal Rule of Criminal Procedure 8(a). Id. We then review the denial of a motion to sever under Federal Rule of Criminal Procedure 14 for abuse of discretion. Id.

ii. Analysis

Joinder of offenses is proper where they “are of the same or similar character ... or are connected with or constitute parts of a common scheme or plan.” Fed.R.Crim.P. 8(a). We construe *1337 this requirement broadly, in favor of initial joinder. See Hersh, 297 F.3d at 1241. We look only to the allegations stated on the face of the indictment to make our determination. United States v. Diaz, 248 F.3d 1065, 1102 (11th Cir.2001).

Barsoum argues joinder was improper because “the conspiracy [was] substantially different than the substantive counts” and “because the charged offenses were not connected to a common scheme or plan.” Appellant’s Br. at 47. For instance, he argues some of the events comprising the conspiracy took place earlier in time and at a different pharmacy than the substantive distribution acts. Id. Bars-oum’s argument is meritless. Looking only to the Superseding Indictment, it is obvious that the substantive distribution counts (distributing Oxycodone not for a legitimate medical purpose and outside the course of professional practice) are similar in character to his conspiracy count (conspiracy to distribute Oxycodone not for a legitimate medical purpose and outside the course of professional practice). The fact that Barsoum also distributed pills at a later date or pharmacy is irrelevant.

We turn now to the second step of the analysis. We will reverse the denial of Barsoum’s motion to sever only if he “demonstratefs] that he received an unfair trial and suffered compelling prejudice. This is a heavy burden ... which mere conclusory allegations cannot carry.” Slaughter, 708 F.3d at 1213 (quoting United States v. Walser, 3 F.3d 380, 386 (11th Cir.1993)).

Barsoum argues the District Court abused its discretion in refusing to sever the conspiracy counts from the substantive counts because: (1) the refusal to sever prejudiced him by ultimately preventing him from advancing an entrapment defense; 11 (2) the jury may have used evidence of the substantive counts to infer guilt of the conspiracy count, and vice versa; and (3) the jury may have cumulat-ed the evidence to find overall guilt. Appellant’s Br. at 49-51.

First, Barsoum’s argument that the refusal to sever ultimately prevented him from advancing an entrapment defense is premised only on the unfounded and con-clusory assertion that, had the counts been severed, “evidence of the conspiracy ... would not have clouded the court’s judgment.” Appellant’s Br. at 49-50. Such a conclusory assertion fails to meet the heavy burden of demonstrating an unfair trial and compelling prejudice. Slaughter, 708 F.3d at 1213.

As to Barsoum’s second argument, regarding potential jury confusion, it is based on hypothetical (not proven) jury confusion. Furthermore, evidence of his substantive distribution was relevant evidence of his conspiracy, so it was necessary for the jury to consider such evidence when deciding his culpability for the conspiracy.

As to Barsoum’s third argument, it is also speculative in that he only contends the jury may have cumulated evidence to improperly infer guilt. We reject such speculation. The District Court properly instructed the jury to consider each count separately, and we presume the jury followed these instructions. See United States v. Williams, 526 F.3d 1312, 1321 (11th Cir.2008). There is nothing in the record to suggest otherwise.

Having addressed each of Barsoum’s arguments on this issue, we find no abuse of *1338 discretion in the District Court’s denial of Barsoum’s motion to sever.

F. Denial of Barsoum’s Rule 801(d)(2)(E) Motion

The District Court denied, without explanation, Barsoum’s pretrial motion to exclude the portion of Stevens’ testimony that the government offered as coconspirator statements under Federal Rule of Evidence 801(d)(2)(E). 12 Barsoum then objected on the same basis to that portion of Stevens’ testimony during trial, but the District Court overruled his objection, without explanation.

At trial, the government asked Stevens about the scheme at St. George’s Pharmacy wherein he could bring prescriptions under any patient name and have them filled. Specifically, the government asked how Stevens knew that Wahba understood that the prescriptions Stevens had been providing him at St. George’s were fake. 13 Stevens responded: “We have spoken quite a few times on changing names [for the prescriptions], [to] get different names in.” Stevens later expounded on the scheme stating Wahba had a list of names, but that at some point “[they] had to discontinue using these names and [] come up with some new names.”

Barsoum argued in his pretrial motion that any evidence the government sought to admit as coconspirator testimony would be inadmissible at trial until the government proved a single conspiracy involving the coconspirator at issue. The government responded that the evidence at trial would amply demonstrate that Barsoum was involved in a conspiracy to illegally distribute Oxycodone, that Barsoum had conversations with his coconspirators, and that some of these conversations would be coconspirator statements admissible under Rule 801(d)(2)(E).

Barsoum argued in his trial objection that Stevens’ testimony, specifically, was inadmissible because the government had not yet proved a conspiracy. The government responded that it had established a conspiracy, implicitly arguing that Wahba and Barsoum were coconspirators.

Here, Barsoum argues that Wahba was the out-of-court declarant in Stevens’ testimony, and that the testimony is inadmissible under Rule 801(d)(2)(E) because there was no conspiracy between Wahba and Barsoum. In other words, Barsoum argues that the testimony was inadmissible as coconspirator statements because Wah-ba and Barsoum were not coconspirators. The government responds that any error in admitting Stevens’ testimony was harmless, given the “plethora” of evidence that Barsoum had interacted with Oxycodone dealers and provided them names and other information to create fake prescriptions.

i. Standard of Review

We review the District Court’s ev-identiary rulings for abuse of discretion. General Elec. Co. v. Joiner, 522 U.S. 136, 141, 118 S.Ct. 512, 517, 139 L.Ed.2d 508 (1997). Additionally, we may affirm for any reason supported by the record, even if not relied upon by the District Court. United States v. Hall, 714 F.3d 1270, 1272 (11th Cir.2013).

ii. Analysis

A coconspirator’s out-of-court statements may be admitted against a defendant if the government shows by a preponderance of the evidence that there was a conspiracy involving the declarant and the defendant, and the statement was made *1339 during the course and in furtherance of the conspiracy. Fed.R.Evid. 801(d)(2)(E); United States v. Hasner, 340 F.3d 1261, 1274 (11th Cir.2003); United States v. Turner, 871 F.2d 1574, 1581 (11th Cir.1989). Such statements are not hearsay and are therefore not excluded by the rule against hearsay. Fed.R.Evid. 801(d).

We affirm the District Court’s decision to admit this portion of Stevens’ testimony because it did not relay any out-of-court statement. See Hall, 714 F.3d at 1272. Stevens’ first statement was about “his involvement with Wahba,” Appellant’s Br. at 52, and only conveyed that he had spoken with Wahba quite a few times about the subject matter of changing names. Stevens’ second statement merely describes the course of action they took to avoid suspicion from the authorities. Stevens’ testimony references the existence of conversations with Wahba, but does not convey an out-of-court assertion. See Fed. R.Evid. 801(a). Instead, Stevens testified only to his personal knowledge; namely, what led Stevens to believe that Wahba understood the prescriptions were fraudulent.

We also disagree with Barsoum’s argument that the government had not sufficiently established a conspiracy involving Wahba and Barsoum as coconspirators. The District Court heard evidence that: (1) Barsoum is Wahba’s cousin; (2) Wahba passed Oxycodone and Oxycontin pursuant to fake prescriptions at St. George’s Pharmacy before Barsoum began working there; (3) Barsoum worked at St. George’s under Wahba; (4) Barsoum accepted fake prescriptions and large amounts of cash in exchange for large numbers of pills at St. George’s; and (5) Barsoum conferred with his coconspirators in several sub-schemes to use certain fake or stolen patient names and change other information on the prescriptions he accepted. Thus even if Stevens’ testimony conveyed out-of-court assertions by Wahba, we would find no abuse of discretion because the evidence showed that there was a conspiracy involving Wah-ba and Barsoum, and that Wahba’s statements to Stevens were made in furtherance of that conspiracy. We therefore need not address whether the admission of Stevens’ testimony was harmless error.

Because Stevens’ testimony was admissible without resort to 801(d)(2)(E), we affirm its admission as nonhearsay.

G. Denial of Barsoum’s Motion for Mistrial Based on Prosecutor’s Response to his 801(d)(2)(E) Objection

When responding to Barsoum’s 801(d)(2)(E) objection described above, the prosecutor stated in the presence of the jury:

“Your Honor, at this — at this time I think we pretty squarely established a conspiracy, an agreement between these individuals, uh, changing names, changing DEA numbers. It’s obviously in furtherance of that conspiracy.”

D.C. Docket Entry 236, Trial Tr. (Aug. 15, 2012) at 149. Barsoum argued at trial that the prosecutor’s remark conveyed a legal statement to the jury, and that the District Court validated the remark when the Court overruled Barsoum’s objection in front of the jury. Barsoum argues here that the prosecutor’s statement was so prejudicial that it shifted the burden to Barsoum to disprove the existence of a conspiracy, and that no instruction could cure this prejudicial effect.

When Barsoum moved for a mistrial on the basis of the prosecutor’s statement, the District Court denied his motion and gave a lengthy curative instruction, which had been formulated in large part by Bars-oum’s counsel. The instruction stated:

*1340 Earlier today the defense objected to the introduction of certain testimony, and the Government responded that in its view there was a conspiracy proven and I ruled.

You are instructed that it will be up to you to decide whether or not a conspiracy exists in this case. Everybody understands that? That’s your decision to make whether or not a conspiracy exists in this case.

The questions and the objections of the lawyers, as I told you earlier, that is not evidence. The lawyers’ questions are not evidence, the lawyer’s,objections are not evidence. Only the witnesses’ answers are evidence.

And it is up to you again to decide whether a conspiracy exists in this case. And you need to make your decision based on the evidence and not on what the lawyers say.

So you are instructed to follow that request and to follow the instructions that I gave earlier. And I’ll repeat right now, the lawyers’ questions are not evidence, that lawyers’ objections are not evidence.

D.C. Docket Entry 236 at 192-93. Bars-oum contends that, despite this instruction, there is a reasonable probability the outcome of his trial would have been different absent the prosecutor’s statement.

i. Standard of Review .

We review for abuse of discretion the denial of a motion for mistrial based on the prejudicial statements of a prosecutor. United States v. Newsome, 475 F.3d 1221, 1227 (11th Cir.2007).

ii. Analysis

A defendant must show substantial prejudice to be granted a mistrial. United States v. Soto, 399 Fed.Appx. 498, 502 (11th Cir.2010). Substantial prejudice occurs where there is a reasonable probability that, without the remarks, the result of the trial would have been different. Newsome, 475 F.3d at 1227. However, “when a district court gives a curative instruction, [we] will reverse only if the evidence is so highly prejudicial as to be incurable by the [instruction].” Id.

We cannot conclude that the prosecutor’s remark was so highly prejudicial that it could not be cured by an instruction. See id. This is the exact circumstance for which curative instructions are intended. Especially where, as here, the defendant helped formulate the instruction given, and where that instruction specifically identified the prosecutor’s controversial statement and reminded the jury that it was their decision whether or not a conspiracy had been proved, the instruction cured any prejudice. See id. Because the instruction eliminated any reasonable probability that the result of the trial would have been different, we affirm. Id.

H. The Denial of Barsoum’s Motion for New Trial on the Basis of New Evidence

Barsoum moved for a new trial on the basis of “newly-discovered” evidence. The District Court granted him a hearing to present the testimony of three inmates who had purportedly gleaned new evidence from conversations in prison with Scott and Stevens. The first inmate told the District Court (1) that Stevens told him that at one point he had to threaten Bars-oum with a gun to make him continue in the conspiracy, and (2) that Scott may have testified untruthfully because of the prospect of a shortened sentence. 14 The *1341 second inmate stated (1) that Scott told him Barsoum was unaware their scheme was fraudulent and (2) that Scott said Stevens had to threaten Barsoum. The third inmate said that Stevens admitted that he owned the guns found at Barsoum’s arrest and also that he had to pressure Barsoum. The District Court found that all of the “new” evidence was merely impeachment evidence or cumulative of other evidence presented at trial, and therefore denied Barsoum’s motion.

i. Standard of Review

We review for the denial of a motion for new trial on the basis of newly-discovered evidence only for abuse of discretion. United States v. Johnson, 596 F.2d 147, 148 (5th Cir.1979); United States v. Spellissy, 346 Fed.Appx. 446, 451 (11th Cir.2009).

ii. Analysis

To succeed on a motion for new trial based on newly discovered evidence, Barsoum must prove that:

(1) The evidence was discovered after trial, (2) the failure of the [Appellant] to discover the evidence was not due to a lack of due diligence, (3) the evidence is not merely cumulative or impeaching, (4) the evidence is material to issues before the court, and (5) the evidence is such that a new trial would probably produce a different result. United States v. Jernigan, 341 F.3d 1273, 1287 (11th Cir.2003) (citations omitted). Courts should use “great caution” in granting such motions as they are highly disfavored. Id.

We find no abuse of discretion in the denial of Barsoum’s motion. The District Court was in the best position to assess the testimony at its hearing. It determined that the purportedly “newly-discovered” evidence was either impeachment evidence or cumulative of other evidence presented at trial. The District Court fully reasoned through this evidence and the trial record to make its determination. Accordingly, we find no abuse of discretion.

III. CONCLUSION

For the foregoing reasons, we affirm the District Court on all issues presented on appeal.

AFFIRMED.