United States v. Adelfo Pamatmat

NOT RECOMMENDED FOR PUBLICATION File Name: 18a0583n.06

No. 17-1611 FILED UNITED STATES COURT OF APPEALS Nov 26, 2018 FOR THE SIXTH CIRCUIT DEBORAH S. HUNT, Clerk

UNITED STATES OF AMERICA, ) ) ON APPEAL FROM THE Plaintiff-Appellee, ) UNITED STATES DISTRICT ) COURT FOR THE EASTERN v. ) DISTRICT OF MICHIGAN ) ADELFO PAMATMAT, ) OPINION ) Defendant-Appellant. )

BEFORE: THAPAR, BUSH, and NALBANDIAN, Circuit Judges.

JOHN K. BUSH, Circuit Judge. Adelfo Pamatmat, a medical doctor, was convicted of conspiracy to distribute and possess with intent to distribute controlled substances in violation of 21 U.S.C. §§ 841(a)(1), (b)(1)(C) and 846 (Count I), and of health care fraud conspiracy in violation of 18 U.S.C. §§ 1347 and 1349 (Count II). Following his convictions, Pamatmat was sentenced to 228 months of imprisonment on Count I and a concurrent term of 120 months on Count II.

Pamatmat now appeals the district court’s denial of his new trial motion and challenges his sentences as procedurally unreasonable. For the reasons that follow, we AFFIRM.

I. FACTS

A. The Conspiracy

Pamatmat was employed from 2007 to 2009 at a corporation called Compassionate Doctors, P.C. (“Compassionate”). Although serving a small number of legitimate patients, *2 No. 17-1611, United States v. Pamatmat Compassionate mostly engaged in prescribing controlled substances and expensive non-controlled substances to patients who did not need them, and its doctors sometimes performed unnecessary home visits for the purpose of prescribing these substances. Compassionate billed Medicare over $10 million for its fraudulent services. The pharmacies that filled the prescriptions also billed insurance companies and retained the payments.

Pamatmat was involved in the illicit activities of Compassionate and a similar group, Visiting Doctors of America (“VDA”), through signing prescriptions for controlled and expensive non-controlled substances and signing patient charts for patients he did not examine. Pamatmat was paid for signing the charts. In addition, he signed a number of blank prescription pads— without a patient name or a medication name—and sold them to another employee of Compassionate, an unlicensed medical-school graduate named Javar Myatt-Jones. Myatt-Jones filled in the pads with prescriptions for narcotics, which he sold on the street after he obtained them.

Some of Pamatmat’s activities continued after he left the employment of Compassionate. He continued to sign charts for patients he had never seen and to sign blank prescription pads and sell them to Myatt-Jones, who had gone to work for VDA. Pamatmat also referred the patients of “street marketers” to several home health care companies, which billed Medicare for physical therapy that was never performed. Pamatmat was paid for these referrals as well.

In early 2009, the Drug Enforcement Administration (“DEA”) began investigating Compassionate’s and VDA’s (among other companies’) activities. The investigation led officers to Pamatmat, who in December 2010 gave a Mirandized statement confessing to his participation in the conspiracy. In his statement, Pamatmat admitted that he signed “approximately 30 charts and 20 [blank] prescriptions” per month for Myatt-Jones; that most of his prescriptions were for *3 No. 17-1611, United States v. Pamatmat OxyContin 80 milligrams, although “he felt that there was no medical justification” for prescribing it; that for each patient he signed two prescriptions (one for a controlled substance that a marketer, rather than a legitimate patient, would use to obtain the drugs and then sell them on the street, and one for “maintenance drugs” for which a conspirator pharmacy could bill Medicare, regardless of whether the pharmacy actually filled the prescription); “that he knew his activities were illegal;” that, although he did not profess to know for certain what the marketers were doing with the narcotics prescriptions, “[h]e felt that they were selling [the drugs obtained through the prescriptions] on the street;” and that he was paid fifty dollars per home visit for those patients he did see. (R. 1466, Page ID## 14574–79.)

B. The Trial

Pamatmat was indicted on charges of conspiracy to distribute and possess with intent to distribute controlled substances in violation of 21 U.S.C. §§ 841(a)(1), (b)(1)(C) and 846 (Count I), and of health care fraud conspiracy in violation of 18 U.S.C. §§ 1347(a) and 1349 (Count II). He entered a plea of not guilty and his case was tried to a jury.

At Pamatmat’s trial, former DEA task force agent Patrick DeBottis testified to the contents of Pamatmat’s confession. Among the other evidence presented by the government was an undercover video, recorded by Myatt-Jones (who had agreed to cooperate with the government), of Pamatmat signing blank prescription forms and falsified patient charts and stating that he had never seen many of his patients. Also admitted into evidence was a recording made by Myatt-Jones of a phone conversation with Pamatmat in which Pamatmat admitted he had signed prescriptions for patients he had not seen. In addition, two marketers testified about Pamatmat’s involvement in writing prescriptions for controlled substances. The government also presented physical evidence seized at Pamatmat’s home, including patient charts, lists of patient names for *4 No. 17-1611, United States v. Pamatmat home visits, and printouts from the Michigan Automatic Prescription Service (“MAPS”), a prescription-data-compilation system that records the prescriptions for controlled substances filled by all Michigan pharmacies. MAPS contains records of which doctor authorizes each prescription.

The government presented several exhibits summarizing data obtained from MAPS and from STARS, a database of Medicare claim information. Scott O’Connell, the analyst who prepared the exhibits, testified that one of them summarized MAPS data on the number of dosage units prescribed by Pamatmat. These data showed that pharmacies had filled over 3.4 million dosage units of Pamatmat prescriptions for controlled substances between January 2008 and January 2013. Although Pamatmat’s trial counsel, Arthur Landau,1 cross-examined O’Connell on the MAPS exhibit, Landau did not have the underlying data from the MAPS system. When Landau attempted to cross-examine O’Connell on certain aspects of the raw data, the trial judge paused the examination to remind the jury that Landau’s adverting to information about the raw data was not evidence.

One of Pamatmat’s co-conspirators, Dr. Ravi Iyer, testified that only about five percent of Compassionate’s patients had a legitimate need for the controlled substances prescribed for them. The government also presented expert testimony from Dr. Eugene Mitchell, who testified that presigning prescription pads was “not remotely related to any legitimate medical practice.” (R. 1454, Page ID# 12986.)

In an attempt to discredit the government’s evidence as to the amount of illegal activity he engaged in, Pamatmat presented testimony that he worked “full-time” at an emergency room in Caro, Michigan from December 2007 to mid-2010. (R. 1458, Page ID# 13865.)

1 Pamatmat was also represented by Kevin Landau. We refer to the Landaus collectively as Pamatmat’s “trial counsel” in analyzing his ineffective-assistance-of-counsel claim.

No. 17-1611, United States v. Pamatmat

Pamatmat also called a handwriting expert to attempt to show that some of the signatures on prescription pads and patient files that the government presented were forged. The expert, Wendy Carlson, was permitted to give her opinion only as to nineteen signatures (of thirty-two total that she examined when preparing for trial). Carlson’s testimony was limited to those nineteen signatures as a result of a stipulation reached by the government and Pamatmat’s trial counsel. Because Pamatmat’s trial counsel had disclosed Carlson’s credentials and her report over twenty-four hours past the court’s deadline, all of her testimony could have been excluded under Federal Rule of Criminal Procedure 16. Notwithstanding this, the government agreed not to challenge Carlson’s testimony as to the nineteen signatures in exchange for Pamatmat’s trial counsel’s agreement not to argue to the jury the government’s failure to retain its own handwriting expert. However, the government still sought, and the court granted, exclusion of Carlson’s testimony as to all but the first nineteen signatures, because Pamatmat’s trial counsel had disclosed Carlson’s opinions as to the remaining signatures even later: roughly eighteen hours before trial. The court found that the last-minute batch of opinions was “too late to be acceptable.” (R. 1458, Page ID# 13724.)

In preparing for trial, Carlson examined copies of the signed documents instead of originals. At trial, she testified that in her opinion, each of the nineteen signatures she had examined was a forgery. On cross-examination, the prosecutor showed Carlson videos of Pamatmat appearing to sign prescription pads and patient charts; some of those documents corresponded to copies that Carlson had examined and declared were forgeries. Carlson responded in the negative to the prosecutor’s question whether seeing the video changed her opinion about the validity of the signatures. By contrast, “Myatt-Jones testified that he had never forged [Pamatmat’s] signature on a prescription.” (R. 1648, Page ID# 16707.) *6 No. 17-1611, United States v. Pamatmat

At the conclusion of the trial, the jury convicted Pamatmat of both Counts. On Count I, the jury found Pamatmat guilty of prescribing illegally Schedule II, III, IV, and V drugs.2

C. Post-Trial Proceedings

Pamatmat moved for judgment of acquittal and for a new trial. Subsequently, Pamatmat’s trial counsel withdrew, and new counsel was appointed to represent him. Through his new attorney, Pamatmat then filed a second motion for a new trial. In addition, he filed objections to the United States Probation Office’s calculation of his Guidelines sentencing range. The district court denied the first new trial motion and upheld the Guidelines calculation. Then the district court sentenced Pamatmat to 228 months of imprisonment on Count I and a concurrent term of 120 months on Count II. These sentences corresponded to the Guidelines ranges for Pamatmat’s two convictions.

In sentencing Pamatmat on Count I, the district court considered Iyer’s testimony that only about five percent of Compassionate’s practice of patient visits for the purpose of prescribing controlled substances was legitimate. Based on this evidence, the government’s sentencing memorandum—the recommendations of which the district court adopted—made an “extremely conservative estimate” of Pamatmat’s illegal controlled substances prescriptions by doubling the estimated percentage of legitimate patient visits to ten percent. (R. 1553-7, Page ID# 15832.) The district court therefore determined that ninety percent of Pamatmat’s controlled substances prescriptions for the relevant time period should be considered for sentencing purposes.3 This determination resulted in an offense level of 38 under the Guidelines.

2 According to the verdict form’s classifications, Schedule II drugs include OxyContin, Oxycodone, and Opana; Schedule III drugs include Vicodin, Lortab, and hydrocodone; Schedule IV drugs include Xanax/Alprazolam; and the Schedule V drug is Promethazine with codeine cough syrup. 3 The government’s sentencing memorandum noted that data were unavailable for nine months of the period, from January 2008 to January 2013, during which trial testimony established that Pamatmat was involved in the conspiracy.

No. 17-1611, United States v. Pamatmat

In determining the range for Count II, the district court found that, per the Guidelines, Pamatmat deserved an eighteen-level addition to his offense level because the loss amount from his involvement in the conspiracy was between $2.5 million and $7 million. Pamatmat’s Guidelines offense level for the fraud count was therefore 28 (including enhancements not disputed on appeal). The district court based this determination on evidence of two separate areas of illegal activity. The first area was Pamatmat’s referrals to two home health care companies owned by marketer and co-conspirator Deepak Kumar. These referrals resulted in payments on fraudulent Medicare claims totaling $1,944,830.18 for referrals to one of the companies and $698,002.41 for referrals to the other. The second area of illegal activity was Pamatmat’s patient visits made while he worked at Compassionate. The district court found that at least $900,216.87 in false claims had been based on these visits. Adding this number to the number of fraudulent Medicare claims from the home health care referrals resulted in an estimated $3,543,049.46 in false claims attributable to Pamatmat.

After sentencing, the district court denied both Pamatmat’s second new trial motion and his motion for judgment of acquittal. Pamatmat appealed the denial of his second new trial motion to this court.

On appeal, Pamatmat argues that he is entitled to a new trial under Federal Rule of Criminal Procedure 33 because he received ineffective assistance of trial counsel and was prejudiced as a result, and the district court erred in denying his motion without an evidentiary hearing. Appellant Br. at 16.

Pamatmat also challenges his sentence as procedurally unreasonable. Because the district court relied on the government’s estimate of the number of prescriptions Pamatmat wrote illegally,

Thus, the government stated that the prescription amounts in its sentencing exhibits were “significantly understate[d],” even before doubling the estimated five-percent-legitimate rate. (R. 1553-7, Page ID# 15831.)

No. 17-1611, United States v. Pamatmat and because it relied on fraud amounts that included Medicare claims made after Pamatmat left Compassionate’s employ, Pamatmat argues that the district court’s calculation of his Guidelines sentencing ranges was not supported by the evidence and not reasonably foreseeable. Id. at 33. He asks this court to vacate his sentence and remand for resentencing. Id. at 44.

II. DISCUSSION

Pamatmat was convicted of two charges: (1) conspiracy to distribute and possess with intent to distribute controlled substances and (2) health care fraud conspiracy. The elements of conspiracy to commit drug-related offenses under 21 U.S.C. § 846 are (1) “the existence of an agreement to violate the drug laws” and (2) knowledge of, intent to join, and participation in the conspiracy. United States v. Pearce, 912 F.2d 159, 161 (6th Cir. 1990). “Proof of a formal agreement is unnecessary, and a tacit or material understanding among the parties is sufficient to show a conspiracy.” Id. (citation omitted).

The elements of health care fraud conspiracy under 18 U.S.C. § 1349 are (1) an agreement between two or more persons to (2) “knowingly and willfully execute[], or attempt[] to execute, a scheme or artifice . . . to defraud any health care benefit program; or . . . to obtain, by means of false or fraudulent pretenses, representations, or promises, any of the money or property owned by, or under the custody or control of, any health care benefit program, in connection with the delivery of or payment for health care benefits, items, or services.” United States v. Patel, 579 F. App’x 449, 460 (6th Cir. 2014); 18 U.S.C. § 1347(a). A defendant may be convicted of health care fraud under § 1347 without having “actual knowledge of [§ 1347] or specific intent to commit a violation of [§ 1347].” 18 U.S.C. § 1347(b). *9 No. 17-1611, United States v. Pamatmat

A. Pamatmat’s Rule 33 Motion for a New Trial

We review a district court’s denial of a motion for a new trial under Federal Rule of Criminal Procedure 33 for abuse of discretion. United States v. Arny, 831 F.3d 725, 730 (6th Cir. 2016). An abuse of discretion occurs when the district court “relies on clearly erroneous findings of fact, uses an erroneous legal standard, or improperly applies the law.” Id. (citation omitted). Rule 33 provides that a “court may vacate any judgment and grant a new trial if the interest of justice so requires.” Fed. R. Crim. P. 33(a).

Pamatmat argues that the interest of justice requires a new trial under Rule 33 because he received ineffective assistance of trial counsel. See United States v. Munoz, 605 F.3d 359, 373 (6th Cir. 2010) (“A violation of [the] Sixth Amendment right to effective assistance of counsel clearly meets [the Rule 33] standard.” (citations omitted)). A claim based on ineffective assistance of counsel involves a mixed question of law and fact, which we review de novo. Arny, 831 F.3d at 730. Under Strickland v. Washington, 466 U.S. 668, 688, 692 (1984), an ineffective-assistance claim requires showing both deficiency and prejudice. Counsel’s performance was deficient if it “fell below an objective standard of reasonableness.” Id. at 688. To show prejudice, the defendant must establish “a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” Id. at 694.

Pamatmat argues that his trial counsel were ineffective in the following ways: they did not “investigate and obtain the underlying raw data” from MAPS that the government’s witness, Scott O’Connell, used in preparing exhibits to demonstrate the number of dosage units Pamatmat illegally prescribed; they did not “present a witness to challenge” those data or O’Connell’s findings; they did not present evidence that Pamatmat was in a coma from January 19, 2012 to March 14, 2012, a period during which the government argued Pamatmat engaged in illegal *10 No. 17-1611, United States v. Pamatmat activity; and they failed to comply with Federal Rule of Criminal Procedure 16 and the district court’s scheduling order when they missed the deadline for disclosing Wendy Carlson’s qualifications and conclusions as a handwriting expert. Appellant Br. at 19–20. We find that even if trial counsel were ineffective in one or all of the alleged ways, their ineffectiveness did not prejudice Pamatmat as required by Strickland, and therefore the district court correctly denied Pamatmat’s motion for a new trial. We address each of trial counsel’s alleged inadequacies in turn.

As mentioned, the prejudice standard from Strickland requires “a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” 466 U.S. at 694. Pamatmat has not shown that, if his trial counsel had obtained the raw data underlying the MAPS information that O’Connell used to prepare the government’s exhibits and had hired an expert to examine those data, the jury would have been reasonably likely to have acquitted him of either of the conspiracy charges. The raw data at issue are the prescriptions apparently signed by Pamatmat, as opposed to the information stored in MAPS about each prescription, which includes “the name and address of the person receiving the prescription, the name of the pharmacy filling the prescription, the date the prescription was written and the date it was filled.” Appellant Br. at 23; (see R. 1538-1, Page ID# 15631). Pamatmat’s trial counsel objected to the use of the MAPS information as hearsay, and the district court overruled that objection, finding that the business records exception applied. Therefore, even if his trial counsel had accessed and presented raw data, the jury would still have seen the government’s presentation of compilations of those data, showing that Pamatmat had prescribed, and pharmacies had filled, roughly 3.4 million dosage units of controlled substances from 2008 to 2013. *11 No. 17-1611, United States v. Pamatmat

However, Pamatmat argues that because many of the pharmacies reporting data to MAPS were “allegedly involved in the conspiracy,” the “veracity” of the prescriptions is “call[ed] into question.” Appellant Br. at 23. In particular, Pamatmat argues that had his trial counsel obtained the prescriptions, along with an expert to examine them, they might have been able to show that Pamatmat’s signatures had been forged on the prescriptions. Id. at 24. But Pamatmat does not give us any reason to believe that those signatures may have been forged, other than Carlson’s having identified nineteen copies of signed prescriptions and patient charts as forgeries. Indeed, Pamatmat’s Mirandized confession, which the government introduced at trial, included his admission that he had signed prescriptions and patient charts in furtherance of the conspiracy. Furthermore, the government introduced video evidence of Pamatmat signing prescription pads and patient charts.4

Even had Pamatmat’s trial counsel obtained the prescription forms underlying the MAPS compilations, Pamatmat has not shown a reasonable probability that (1) an expert would have determined any of the signatures on those prescriptions were forgeries; (2) a jury would have believed expert testimony that the signatures were forgeries; (3) the jury would have correspondingly discounted the government’s evidence regarding the legitimacy of the signatures, including Pamatmat’s confession, the video, and Myatt-Jones’s testimony that he had never forged Pamatmat’s signature; and (4) finding that some or all of the signatures had been forged, the jury would have acquitted Pamatmat despite the existence of separate evidence of Pamatmat’s involvement in the conspiracy. Therefore, we find that Pamatmat has not demonstrated a reasonable probability that either obtaining the underlying data or retaining an expert to examine those data would have affected the outcome of his trial.

4 Pamatmat does not argue—nor does he assert that his trial counsel should have argued—that he was not the person in the video or that the video was unreliable evidence in any way.

No. 17-1611, United States v. Pamatmat

Next, Pamatmat claims his trial counsel were ineffective in failing to present evidence that he was in a coma for nearly two months in early 2012 and therefore could not have signed prescriptions during that time (or, indeed, at any time between January 16 and December 31, 2012, since Pamatmat argues that he was “physically unable to work” all year). Id. at 23. He argues that his attorneys should have objected to the government’s introduction of MAPS data based on prescriptions filled during this time period. Id. at 24, 26. In addition, he asserts that his attorneys should have argued, based on Pamatmat’s having been ill, that any prescriptions purportedly signed by him during 2012 were forgeries. Id. at 24.

But Pamatmat again fails to show prejudice because his having been sick while prescriptions signed in his name continued to be filled is consistent with evidence that he pre-signed blank prescription forms and sold them to Myatt-Jones. Myatt-Jones testified before the jury that Pamatmat pre-signed blank forms. Therefore, even had the jury known that Pamatmat was sick in 2012, Pamatmat has not shown a reasonable probability the jury would have concluded that no genuine Pamatmat-signed prescriptions could have been filled during 2012. Even had the jury concluded that, moreover, Pamatmat has not shown that the outcome of his trial would have been different. Without the evidence of 2012 activity, the jury still would have heard evidence of several years’ involvement in a conspiracy, which would have been sufficient to convict Pamatmat of the charges.

However, Pamatmat argues that the inclusion of 2012 prescriptions in the MAPS compilations indicates the general unreliability of the MAPS data. Id. at 26. To the extent Pamatmat may be using this argument to reiterate that his counsel should have obtained the underlying prescription forms and hired an expert to examine them, he still has not shown prejudice because he does not present any evidence to indicate that the data are unreliable. He simply asserts *13 No. 17-1611, United States v. Pamatmat that the Pamatmat signatures on 2012 prescriptions were forged. But as the government points out, far fewer Pamatmat-signed prescriptions were filled in 2012 than in the preceding two years. The MAPS compilations showed that 1,399,836 dosage units were filled based on Pamatmat prescriptions in 2010; 1,713,526 in 2011; and 343,457 in 2012. Appellee Br. at 17–18. Such a reduction is consistent with Pamatmat’s having been unable to sign new prescriptions in 2012, meaning the other conspirators would have had to use the limited number of pre-signed prescription forms that Pamatmat had sold before his hospitalization. Therefore, Pamatmat has failed to show a reasonable probability the jury would have acquitted him of the conspiracy charges had his attorneys presented evidence that he was sick and did not work in 2012.

Finally, Pamatmat argues he was prejudiced by his trial counsel’s failure to retain a handwriting expert in a timely fashion and to disclose the expert’s qualifications and report to the court before the deadline.5 Appellant Br. at 26. Pamatmat asserts that his trial counsel did not hire Carlson “until the last week of the jury trial.” Id. As a result, she had time to examine only thirty-two signatures. Pamatmat argues that “[h]ad [c]ounsel . . . timely hired a handwriting expert, she could have analyzed every single exhibit offered by the Government during the multi-week trial.” Id. at 27.

Because his argument that his signature had been forged on prescription pads and patient charts was “[a] cornerstone” of his defense, Pamatmat also argues that he was prejudiced when his trial counsel were forced—by their tardiness in retaining an expert—to strike an agreement with

5 Pamatmat argues that his trial counsel “initially violated [Rule] 16’s deadlines, and subsequently violated the [district] [c]ourt’s deadline of Friday at 4:00 p m. to provide Ms. Carlson’s report . . . .” Appellant Br. at 27. Rule 16(b)(1)(C), governing the defendant’s disclosure of expert qualifications, opinions, and bases for opinion, does not establish a deadline. However, the district court had set deadlines in this case that Pamatmat’s trial counsel failed to meet.

No. 17-1611, United States v. Pamatmat the government that precluded Pamatmat from arguing that the government had not retained its own handwriting expert to testify to the signatures’ authenticity. Id. at 26; see also id. at 27.

To prove prejudice, however, Pamatmat needs to show that, had his expert been permitted to examine and testify about a greater number of signatures than the nineteen she actually testified about, the jury would have been reasonably likely to acquit. See Strickland, 466 U.S. at 694. Pamatmat argues that his “defense was largely based on the fact that any involvement in the drug conspiracy . . . was due to his signature being forged.” Appellant Br. at 28. If that is true, though, Pamatmat fails to explain how an expert’s testifying that more than nineteen of the signatures had been forged would have altered the jury’s conclusion about his involvement in the conspiracy. After all, the jury did hear Pamatmat’s expert say she thought all nineteen of the signatures she reviewed had been forged. The jury may have inferred from that testimony that many more of Pamatmat’s alleged signatures were forgeries, or it may simply have accepted Carlson’s opinion as to the nineteen. Alternatively, it may not have believed her at all. In any event, the jury did not find that Carlson’s testimony outweighed Pamatmat’s confession, the video of Pamatmat signing prescription pads and patient charts, and Myatt-Jones’s testimony that he had never forged Pamatmat’s signature. And Pamatmat has not made any arguments indicating why Carlson’s testifying as to the inauthenticity of more signatures would have weighed more heavily with the jury.6 As to the government’s failure to retain its own handwriting expert, the jury could see that the government had no expert regardless of whether Pamatmat’s counsel stressed the point in

6 The government asserts that Pamatmat’s expert was “simply not a credible witness” because of the nature of her credentials and training, so her testimony as to more signatures would not have helped Pamatmat’s defense. Appellee Br. at 18–19. We agree that Pamatmat does not show that more Carlson testimony would have aided him, but we also note that Pamatmat does not argue his counsel should have retained a different expert who might have been more credible. Therefore, we make no comment on Carlson’s credibility beyond noting that the jury did not appear to find her more credible than the various pieces of evidence the government presented to demonstrate the validity of Pamatmat’s signatures.

No. 17-1611, United States v. Pamatmat argument. Pamatmat offers us no reason to believe that his counsel’s inability to argue the point was reasonably likely to have influenced the outcome of his trial.

Although we find that none of counsel’s alleged errors prejudiced Pamatmat, Pamatmat argues that the district court should have “consider[ed] the combined effect of all acts of counsel found to be constitutionally deficient, in light of the totality of the evidence in the case,” to determine whether all of his counsel’s allegedly deficient acts combined to prejudice him. Id. at 29 (quoting United States v. Dado, 759 F.3d 550, 563 (6th Cir. 2014)). Pamatmat argues that the district court “summarily” disposed of his new trial motion and request for an evidentiary hearing on the motion. Id.

Summary or not, the district court’s analysis directly addressed Pamatmat’s contention that his counsel’s alleged errors, taken as a whole, prejudiced his defense. After addressing each of Pamatmat’s arguments regarding his counsel’s alleged individual errors, the district court stated:

To justify the extraordinary remedy of a new trial based on cumulative error, Defendant must show that . . . the aggregate effect of the “individually harmless errors was so prejudicial as to render his trial fundamentally unfair.” Given the weight of the evidence against Defendant, including his oral confession, the court is not persuaded that his retained trial counsel’s arguably sub-optimal performance actually prejudiced Defendant to that degree.

(R. 1648, Page ID# 16708 (emphases added) (quoting United States v. Trujillo, 376 F.3d 593, 614 (6th Cir. 2004)).). Although we must review the district court’s ineffective-assistance analysis de novo, we do not require the district court to state its conclusion at any particular length. If we are satisfied that the district court’s conclusion was correct, we must affirm.

We are satisfied. Although Trujillo, which the district court cited to establish the standard a defendant must satisfy to show cumulative prejudicial error, was a case about alleged errors by a court, not trial counsel, the Strickland standard for assessing cumulative error is not materially different. See Strickland, 466 U.S. at 696 (“Taking the unaffected findings as a given, and taking *16 No. 17-1611, United States v. Pamatmat due account of the effect of the errors on the remaining findings, a court making the prejudice inquiry must ask if the defendant has met the burden of showing that the decision reached would reasonably likely have been different absent the errors.”); see also id. at 686 (stating that the purpose of the constitutional requirement of effective assistance of counsel is “to ensure a fair trial”); Stewart v. Wolfenbarger, 468 F.3d 338, 361 (6th Cir. 2006) (“When determining prejudice, the [c]ourt must consider the errors of counsel in total, against the totality of the evidence in the case.”).

Furthermore, we find that the district court reached the correct conclusion under Strickland. Considered in the aggregate, the alleged errors of Pamatmat’s trial counsel were not reasonably likely to have affected the outcome of his prosecution. Suppose trial counsel had done everything Pamatmat claims they should have done: obtain the raw data underlying the MAPS compilations of Pamatmat prescription information, hire an expert to analyze those data, present evidence to show that Pamatmat was sick in 2012 and did not work, and comply with the district court’s deadline for disclosure of the handwriting expert’s qualifications and report so they could argue that the government had failed to retain its own handwriting expert. Even then, the jury would still have been presented with evidence of Pamatmat’s confession, with evidence that Pamatmat pre-signed prescription forms to be filled in later, with Myatt-Jones’s testimony that he never forged any of Pamatmat’s signatures on those pre-signed forms, with Dr. Ravi Iyer’s testimony that only about five percent of Compassionate’s patients were legitimate, with Dr. Eugene Mitchell’s testimony that signing blank prescription forms was unrelated to legitimate medical practice, and with the rest of the proof the government presented at trial. Pamatmat does not argue that he has learned anything about the MAPS data suggesting they are unreliable; nor does he now present any evidence suggesting the Pamatmat prescriptions from 2012 were not attributable to *17 No. 17-1611, United States v. Pamatmat pre-signed prescription forms. And he does not give us any basis to believe that his handwriting expert’s testimony as to more than nineteen signatures would have influenced the jury in his favor. Consequently, we do not find it reasonably probable that, had trial counsel conducted themselves in the ways Pamatmat suggests, he would have been acquitted of either conspiracy charge.

Pamatmat argues prejudice from his trial counsel’s alleged deficiency by relying on Arny, where we affirmed the district court’s grant of a new trial, based on ineffective assistance of trial counsel, to a doctor who had been convicted of similar crimes to those at issue here. 831 F.3d at 736. But this case is not like Arny. In Arny, the defendant’s counsel had made misrepresentations to him, including about the availability of a potentially key witness who could have testified that Arny took over her practice and was engaged in legitimately prescribing controlled substances. Id. at 729. Arny had “repeatedly asked trial counsel to call” this witness, but they had not done so, and they had not even interviewed her to determine whether her testimony would be valuable to Arny’s defense. Id. at 729, 732–33. In addition, Arny’s counsel had failed to meet with one of their other witnesses before trial and had presented the testimony of one witness who made several statements on the stand that were directly prejudicial to Arny’s defense. Id. at 729–30. Furthermore, counsel failed to present testimony by any of Arny’s legitimate patients; within a short time after trial, Arny secured affidavits from six of these patients, who were prepared to testify to their real need of the controlled substances he had prescribed for them. Id. at 730. Meanwhile, the government had presented testimony by four of Arny’s patients, who testified that they had not needed the drugs he prescribed for them. Id. at 729. The government also presented the testimony of one expert who opined that Arny’s practice was illegitimate, and it called ten additional witnesses. Id. Arny’s attorneys called only three witnesses, including Arny himself. Id. *18 No. 17-1611, United States v. Pamatmat

By contrast, Pamatmat does not allege that his trial counsel failed to communicate with him regarding key aspects of his defense or failed to follow his instructions about investigating particular meritorious avenues of argument. Although he argues his trial counsel should have obtained the data underlying the MAPS calculations, he does not assert that he asked them to do so during the trial process or give any reason beyond speculation for believing obtaining the data would have helped his defense. Neither does he assert that his trial counsel overlooked key witnesses (other than, as he argues, an expert to examine the MAPS data) or made misrepresentations to him or to the court. He does not argue his trial counsel did not call enough witnesses or did not call credible witnesses. In no way does he argue that he received inadequate counseling regarding what plea to enter or what defense to assert. And he does not offer us any reason to believe that his trial counsel could have called a witness to testify that Pamatmat was engaged in a legitimate medical practice at Compassionate, and in his work with Myatt-Jones and the other conspirators, and counter Iyer’s testimony that there was nothing legitimate about Pamatmat’s signing blank prescriptions.

Pamatmat’s trial counsel did attempt to discredit the possibility that Pamatmat was engaged in illegal activity for several years during the alleged scheme by presenting testimony that Pamatmat worked “full-time” at an emergency room in Caro, Michigan from December 2007 to mid-2010. (R. 1458, Page ID# 13865.) Although this may seem like the kind of testimony regarding legitimate medical practice that could have helped Pamatmat, it was not inconsistent with the government’s evidence that Pamatmat also worked for Compassionate during part of that time and continued to illegally sign prescription forms and participate in fraudulent referral schemes after leaving Compassionate. *19 No. 17-1611, United States v. Pamatmat

In support of his argument that his counsel should have conducted a more rigorous investigation before trial, Pamatmat points to an affidavit his new counsel obtained from a former FBI Agent, James Hoppe. Appellant Br. at 23. Hoppe’s affidavit states that “Pamatmat did write [prescriptions] to individuals that he examined as an emergency room physician at Caro Hospital.” Id. (citing R. 1538-1, Page ID## 15630–32). However, Pamatmat fails to explain how hearing testimony to this effect would have persuaded the jury that Pamatmat was too busy tending legitimate patients in Caro to be writing fraudulent prescriptions for the conspiracy. He also neglects to mention that after trial, government witness O’Connell analyzed prescriptions Pamatmat wrote during the time he was employed as an ER doctor and found “that none of the Schedule II prescriptions—the most powerful prescriptions sold by the marketers—were filled at pharmacies near Caro . . . .” Appellee Br. at 16–17 (citing R. 1684, Page ID## 17102–03). Presumably, had Pamatmat presented testimony by a witness like Hoppe, the government would have had the opportunity to present testimony along the lines of O’Connell’s post-trial analysis, suggesting that Pamatmat’s Schedule II prescriptions, at least, were all illegitimate. Pamatmat does not persuade us that this exchange of testimony would have aided his defense.

Most importantly, Pamatmat does not assert that his Mirandized confession was defective or subject to being discredited by another hypothetical witness’s testimony at trial.7 He does not explain why the jury would have been likely to find him not guilty after hearing his confession, which included his admission that he signed blank prescription forms and his admission that he knew his conduct was illegal. In sum, Pamatmat has failed to convince us, either by analogy to

7 Pamatmat moved to suppress his confession before trial, but the district court denied the motion, and Pamatmat does not now argue that his trial counsel should have made any additional efforts to mute the effect of government witness DeBottis’s testifying to the contents of that confession at trial.

No. 17-1611, United States v. Pamatmat Arny or in any other way, that his trial counsel’s conduct prejudiced him such that his trial would have been reasonably likely to have ended differently absent their alleged errors.8

Therefore, the district court did not abuse its discretion in denying Pamatmat’s motion for a new trial under Federal Rule of Criminal Procedure 33.

B. Pamatmat’s Challenge to His Sentencing

Pamatmat also challenges his sentencing as procedurally unreasonable. We review the procedural reasonableness of a district court’s sentencing decisions for abuse of discretion.9 Gall v. United States, 552 U.S. 38, 51 (2007). This review requires the following inquiry:

[T]he appellate court . . . must . . . ensure that the district court committed no significant procedural error, such as failing to calculate (or improperly calculating) the Guidelines range, treating the Guidelines as mandatory, failing to consider the [18 U.S.C.] § 3553(a) factors, selecting a sentence based on clearly erroneous facts, or failing to adequately explain the chosen sentence . . . .

Id. We will address first Pamatmat’s challenge to the district court’s determination of drug quantities for his sentence as to Count I and second his challenge to its determination of actual or intended loss for his sentence as to Count II.

1. Drug Quantity Determination

Where the Guidelines range is based on a calculation of drug quantities, we review the district court’s determination of quantity for clear error. United States v. Jeross, 521 F.3d 562, 570 (6th Cir. 2008). “A district court may not ‘hold a defendant responsible for a specific quantity of drugs unless the court can conclude the defendant is more likely than not actually responsible for a quantity greater than or equal to the quantity for which the defendant is being held

8 We hold only that Pamatmat has not established that his trial would have been reasonably likely to have ended in acquittal on either conspiracy charge had his trial counsel conducted his defense in the ways he thinks they should have. Pamatmat does not appear to base his ineffective-assistance claim on any alleged prejudice at sentencing, and he notes that “the jury did not determine any drug quantities.” Appellant Br. at 3 (citing R. 1160, Page ID# 8142). 9 In his brief, Pamatmat also states the Gall standard for review of a sentence’s substantive reasonableness. Appellant Br. at 34. However, he does not argue that his sentence was substantively unreasonable, so we address only the procedural reasonableness of his sentence.

No. 17-1611, United States v. Pamatmat responsible.’” Id. (quoting United States v. Walton, 908 F.3d 1289, 1302 (6th Cir. 1990)). In other words, a district court may estimate the amount of drugs for which a defendant is responsible, as long as a preponderance of the evidence supports the estimate. Id.

Pamatmat argues that the district court improperly calculated his Guidelines range. Appellant Br. at 34. As with his ineffective-assistance claim, Pamatmat argues that the MAPS data regarding how many dosage units of controlled substances he prescribed are “unreliable” because they are not raw data. Id. at 38. Although Pamatmat cites Hoppe’s affidavit to support this claim, the affidavit does not state that the MAPS data are unreliable. It simply states Hoppe’s “belie[f] that this data is [sic] second hand data, not original data or raw data, and was [sic] most likely compiled by someone at the State of Michigan from information received by the State of Michigan from reporting pharmacies.” (R. 1538-1, Page ID# 15631.) Pamatmat fails to explain why the data’s having been compiled from pharmacies’ reports makes them unreliable.

Indeed, this court has previously determined, in response to a challenge by one of Pamatmat’s convicted co-conspirators, that the MAPS data are reliable enough to support sentencing determinations. United States v. Geralt, 682 F. App’x 394, 406–07 (6th Cir. 2017). The co-conspirator, John Geralt, had also argued “that some of the prescriptions attributed to him may have been forgeries” and that the data compilations “lack[ed] trustworthiness . . . because the state prescription database depends on information reported by pharmacies, including those that participated in the conspiracy.” Id. at 407 (internal quotation marks omitted). We found that because Geralt presented no evidence of “actual forgeries or misreported prescriptions . . . it was not clear error for the district court to find that O’Connell’s [the same O’Connell who testified at Pamatmat’s trial] testimony and the state database accurately reflected the number and type of prescriptions written by Geralt.” Id. We find that the same analysis applies in Pamatmat’s case. *22 No. 17-1611, United States v. Pamatmat Although Pamatmat contends that (unlike Geralt) he has presented evidence of actual forgeries through Carlson’s testimony, that testimony was contradicted by Pamatmat’s confession and other government evidence tending to show that Pamatmat had signed many prescription forms illegally. Pamatmat offers no other reason, beyond those Geralt also argued, to doubt the validity of the MAPS data. Therefore, we adopt our reasoning in Geralt with respect to these data’s reliability.

But Pamatmat argues that the district court should not have used an estimate of the percentage of controlled substances that were attributable to illegal, as opposed to legitimate, prescriptions. Appellant Br. at 41. He claims that the district court was wrong to adopt the government’s estimate, which was based on testimony that only five percent of Compassionate’s controlled-substances prescription practice was legitimate. Id. That estimate led the district court to sentence based on the assumption that ninety percent of the controlled substances that Pamatmat prescribed between January 2008 and January 2013 were prescribed illegally. Far from being “a completely arbitrary figure,” id. at 42, this drug amount calculation was based on the government’s conservative estimate of the percentage of Pamatmat’s prescription practice that was illegitimate. Indeed, by adopting the government’s recommendation, the district court did exactly what Pamatmat argues it should have done: it “err[ed] on the side of caution and only h[e]ld the defendant responsible for that quantity of drugs for which the defendant is more likely than not actually responsible.” Id. (quoting United States v. Meacham, 27 F.3d 214, 216 (6th Cir. 1994)). As the prosecutor mentioned at the sentencing hearing, if one considered ninety percent of the prescriptions of oxycodone alone, for the one-year period from 2008 to 2009, that would have produced an offense level of 38. (R. 1684, Page ID# 17103.)

Pamatmat has given us no reason to doubt that the district court made a reasonable and conservative estimate, which our precedent permits, in determining the drug quantities underlying *23 No. 17-1611, United States v. Pamatmat his sentencing. See Jeross, 521 F.3d at 570. Therefore, he has not shown that the district court made a clearly erroneous determination of drug quantities, and he has not shown that the district court abused its sentencing discretion in a way that would render his sentence procedurally unreasonable. See Gall, 552 U.S. at 51.

2. Fraud Loss Amount

Pamatmat also challenges his sentencing for the fraud conviction, arguing that the district court erred in considering loss amounts that occurred after 2009—in other words, after Pamatmat had left Compassionate.10 In reviewing a Guidelines calculation based on a loss amount in a fraud case, we ask whether a preponderance of the evidence supports the district court’s determination of loss, United States v. Wendlandt, 714 F.3d 388, 393 (6th Cir. 2013), and we may affirm if the district court made a “reasonable estimate of actual or intended loss within broad ranges,” United States v. Howley, 707 F.3d 575, 583 (6th Cir. 2013) (citations and internal quotation marks omitted).

Pamatmat argues that “[w]hile Compassionate continued to bill under . . . Pamatmat’s identification numbers” after he left Compassionate’s employ, “there was no way . . . Pamatmat would have known that was what” the Compassionate conspirators were doing. Appellant Br. at 36. Therefore, Pamatmat argues that “it was not foreseeable” that Compassionate would continue to make fraudulent gains by using his identification numbers. Id. at 37. These arguments are unavailing. The government established at trial that Pamatmat continued to engage in fraudulent activity after he left Compassionate. Even excluding the estimated loss amount attributable to

10 The government suggests we should apply plain-error review instead of abuse-of-discretion review to this argument because Pamatmat did not make an objection at sentencing that was “specific enough to afford the district court an opportunity to address the particular claim.” Appellee Br. at 34 (quoting United States v. Simmons, 587 F.3d 348, 356 (6th Cir. 2009)). However, the government does not develop this argument, and we find that the district court did not abuse its discretion, so we need not decide whether the less rigorous plain-error standard of review should apply.

No. 17-1611, United States v. Pamatmat Pamatmat’s work at Compassionate, his fraudulent home health care referrals continued after his employment with Compassionate ended, and the estimated totals from Pamatmat’s home health care referrals alone still resulted in a fraud loss amount of between $2.5 million and $7 million— the amount that resulted in the eighteen-level addition that Pamatmat disputes. (R. 1553-7, Page ID## 15832–33.) As the government points out, these home health care referrals formed “[t]he major component of the fraud calculation,” Appellee Br. at 33, but Pamatmat does not assert that the district court made an error in considering them. Instead, Pamatmat focuses on the post-2009 Compassionate activity of which he claims he was unaware.

But we find it hard to accept Pamatmat’s argument that “it was not foreseeable” that Compassionate would continue to use his billing number to obtain fraudulent prescriptions, Appellant Br. at 37, when he knew he was engaged in illegal activity while employed at Compassionate and presumably knew that only five percent of Compassionate’s patients (at most) were legitimate. We also note that although the district court’s estimate of Compassionate patient visits for which Pamatmat billed controlled-substance prescriptions was based on visits between the years 2007 and 2013—and Pamatmat left Compassionate in 2009—the district court applied a very conservative estimate of the total number of Medicare billings attributable to Pamatmat.

Specifically, the district court relied on the government’s fraud calculation, which estimated that—based on billing number data—Pamatmat had prescribed over twenty-five percent of Compassionate’s controlled-substance prescriptions, stemming from patient visits, between 2007 and 2013. The district court reduced the estimated percentage of patients attributable to Pamatmat to ten percent and then reduced that estimate by another ten percent to account (liberally) for possible legitimate patients. Finally, based on this significantly reduced estimate of the number of illegitimate patient visits attributable to Pamatmat, the district court found Pamatmat responsible *25 No. 17-1611, United States v. Pamatmat for $900,216.87 in fraudulent Medicare claims. (R. 1553-7, Page ID## 15382–33.) The district court then added this number to the totals attributable to Pamatmat’s participation in the fraudulent home health care referral program and determined the result was a fraud loss between $2.5 million and $7 million, making Pamatmat eligible for an eighteen-level addition to his offense level. Pamatmat does not argue that the district court’s method of discounting Pamatmat’s involvement in patient visits was inadequate to deal with the possibility that Pamatmat’s billing number was used after he left Compassionate and without his knowledge.11

Because Pamatmat has not shown that the district court’s estimate of the actual fraud loss was unreasonable, he has not shown that the district court made a clearly erroneous finding of fact, as we must find under Gall to hold that a Guidelines calculation is procedurally unreasonable. See 552 U.S. at 51. We find that a preponderance of the evidence supports the district court’s reasonable estimate of the fraud loss amount. Therefore, the district court did not abuse its discretion in sentencing Pamatmat based on a Guidelines level of 28 on his fraud conviction.

III. CONCLUSION

For the foregoing reasons, we AFFIRM the district court’s denial of Pamatmat’s motion for a new trial and AFFIRM his sentences.

11 It is arguable that even if Pamatmat’s billing number was used by Compassionate without his specific knowledge, he could be held responsible for prescriptions written after he left Compassionate because they were foreseeable as a result of his involvement in the conspiracy. See U.S.S.G. 1B1.3(a)(1)(B). We need not analyze this question because we find that regardless of whether Pamatmat’s billing number was used without his knowledge, the district court did not commit clear error in calculating his Guidelines range.