United States v. Anmy Tran

NOT RECOMMENDED FOR FULL-TEXT PUBLICATION
File Name: 15a0301n.06

Case No. 14-1092
FILED
Apr 27, 2015
DEBORAH S. HUNT, Clerk
UNITED STATES COURT OF APPEALS
FOR THE SIXTH CIRCUIT

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

BEFORE: KEITH, COOK, and DONALD, Circuit Judges.

COOK, Circuit Judge. A jury convicted Defendant-Appellant Anmy Tran, a Michigan

podiatrist, of several offenses arising from her participation in a healthcare-fraud conspiracy

organized by Babubhai Patel, the owner of several pharmacies and home-healthcare companies

in the greater Detroit area. Tran argues that several evidentiary rulings denied her a fair trial and

challenges the procedural reasonableness of her sentence. We AFFIRM the district court’s

judgment.

I.

A jury convicted Tran of participating in the following conspiracy, summarized by a prior

panel of this court:

The scheme to defraud insurers depended on the participation of
physicians, pharmacists, recruiters, and patients. [Babubhai] Patel paid cash
Case No. 14-1092
United States v. Tran

bribes to physicians to entice them to write patient prescriptions for expensive
medications and controlled substances that could be billed to Medicare, Medicaid,
or private insurers through the Patel pharmacies. He paid kickbacks to managers
of health-related companies so that they would send patients to his pharmacies,
and he employed “marketers” to recruit “patients” directly from the streets.

Pharmacists facilitated the criminal activity by charging insurers for
expensive medications that were ordered from wholesale distributors and held in
inventory but not dispensed to patients. These surplus medications were later
returned to the supplier for credit or sold on the black market. Pharmacists also
billed insurers for controlled substances that the pharmacists knew were illegally
prescribed. These controlled medications included hydrocodone (Vicodin,
Lortab), oxycodone (Oxycontin), alprazolam (Xanax), and codeine-infused cough
syrup. When filling prescriptions, the pharmacists usually “shorted” the number
of dosage units placed in the medication vials for patients, billed the insurers for
the full drug quantities prescribed, and then sold the excess pills on the street.

United States v. Patel, 579 F. App’x 449, 451–52 (6th Cir. 2014).

Patel recruited Tran in 2007 to participate in this conspiracy by paying her $50,000. She

used the money to complete the down payment of the office building that housed both her

podiatry practice and Patel’s Highland Park Pharmacy. Patel also paid $1,200 toward each

month’s mortgage payment, $10,000 to cover the building’s 2010 tax bill, and $17,500 to Tran

through various intermediaries.

In exchange, Tran prescribed many of her patients a cocktail of eight to twelve drugs:

one controlled pain medication combined with several non-controlled drugs, requested by Patel

and his pharmacists because of their high profit margins. These patients would then proceed to

Highland Park, where Patel’s pharmacists would dispense the full amount of controlled pain

medication but only half of the non-controlled drugs. Highland Park would then bill the

insurers—Medicaid, Medicare, and Blue Cross Blue Shield—for the entire prescription. Many

patients forced the pharmacists to count out hydrocodone pills and other controlled substances at

-2-
Case No. 14-1092
United States v. Tran

the counter to ensure they received the prescribed amount, though they rarely complained about

not receiving the full amount of non-controlled medication. And Tran always prescribed the

non-controlled medicine in the same dosage, despite some patients reporting a growing surplus.

In addition, Tran accepted patients who were recruited off of the street by other

conspirators. She continued to prescribe controlled pain medication to these patients even after

learning that they were known or suspected “doctor shoppers”: patients recently received the

same type of controlled substance from a different doctor. She also referred a number of

recruited patients to home-healthcare companies operated by Patel without conducting a proper

examination, for which she received a $100 kickback for every patient approved by the insurer.

Patel also paid Tran $500 for every one of her own patients referred to his home-healthcare

services and who were approved by the insurer.

The district court tried Tran jointly with two of her co-conspirators. The jury convicted

Tran of conspiracy to commit healthcare fraud, to distribute controlled substances, and to pay

and receive healthcare kickbacks. Post-trial, the district court denied her motion for a judgment

of acquittal or new trial.

At sentencing, the district court adopted the Presentence Investigation Report’s (PSR)

findings with respect to the amount of loss and drug quantity attributable to Tran. According to

the PSR, Tran’s role in the conspiracy cost the insurers a combined $4,475,193.40.1 This figure

included the cost of every prescription written by Tran and filled at Highland Park

($2,343,498.99), the costs incurred by all patients referred to one of Patel’s home-healthcare

services ($289,858.50), and fifty percent of Tran’s bills for medical services ($1,841,835.99).

This resulted in an adjusted offense level of twenty-eight for the combined offenses of

1
The PSR’s total loss calculation appears to be eight cents short.
-3-
Case No. 14-1092
United States v. Tran

conspiracy to commit healthcare fraud and conspiracy to receive healthcare kickbacks. See

U.S.S.G. § 2B1.1(b)(1)(J). The PSR also counted 213,548 units of hydrocodone—every unit

prescribed by Tran and filled at Highland Park—as unlawfully distributed. As a result, Tran

received an adjusted offense level of twenty-eight on the conspiracy-to-distribute count. See id.

§ 2D1.1(c)(7) (2013).

After incorporating the multiple-count adjustment, the district court calculated Tran’s

total offense level at thirty. See id. § 3D1.4(a). Because she lacked any criminal history, this

resulted in a guidelines range of 97 to 121 months’ imprisonment. After reviewing the

sentencing factors of 18 U.S.C. § 3553, however, the district court varied downward and

sentenced Tran to 60 months.

II.

On appeal, Tran contends that the district court abused its discretion in admitting

statistical evidence comparing her prescriptions to other Michigan podiatrists and testimony by

an expert pharmacist that her prescriptions raised ethical “red flags.” She also challenges her

guidelines range, arguing that the court clearly erred in estimating the drug quantity and amount

of loss attributable to her. We discern no abuse of discretion in the evidentiary rulings and

conclude that any error in calculating her guidelines range proved harmless.

A. Statistical Evidence

Tran first challenges the admission of charts and testimony showing that she prescribed

more than twice as much hydrocodone as any other Michigan podiatrist and controlled

substances to eighty-five percent of her Medicare patients. According to this data, she ranked

first in the state for total prescription drug costs charged to Medicare. Tran unsuccessfully

objected to these charts at trial on the ground that the “probative value [was] substantially

-4-
Case No. 14-1092
United States v. Tran

outweighed by a danger of . . . unfair prejudice, confusing the issues, [and] misleading the jury.”

Fed. R. Evid. 403. She argued that the jury might assume she had no medical reason for

prescribing hydrocodone to her patients based solely on the volume prescribed. She renews this

argument on appeal.

We review evidentiary rulings for abuse of discretion. United States v. Davis, 514 F.3d

596, 611 (6th Cir. 2008). “[W]hen reviewing the balancing determinations required by Rule 403,

this court must maximize the probative value of the challenged evidence and minimize its

potential for unfair prejudice.” United States v. Lloyd, 462 F.3d 510, 516 (6th Cir. 2006); see

also United States v. Layne, 192 F.3d 556, 573 (6th Cir. 1999) (holding that trial courts have

particularly broad discretion under Rule 403). Tran fails to meet her burden under this

deferential standard.

As the government points out, this court has upheld the admission of statistical evidence

under similar circumstances. See, e.g., United States v. Weinstock, 153 F.3d 272, 278 (6th Cir.

1998) (upholding the use of comparative statistics in the prosecution of a podiatrist who charged

insurers for unperformed medical procedures); United States v. August, 984 F.2d 705, 713 (6th

Cir. 1992) (upholding a doctor’s conviction for illegal distribution based, in part, on evidence

that his purchases of Schedule III cough syrup “were much greater than the purchases of the

average doctor, pharmacy or hospital”).

Tran attempts to distinguish this precedent on two grounds. First, she complains that

hydrocodone carries a unique stigma because it receives “an enormous amount of unfavorable

publicity.” She claims that a juror might base the decision to convict on emotion, not evidence.

But we see nothing unfair about any prejudice flowing from the negative publicity surrounding

hydrocodone abuse; the government specifically charged Tran with illegally prescribing

-5-
Case No. 14-1092
United States v. Tran

hydrocodone. See Fed. R. Evid. 403. The relative amount of negative publicity garnered by the

drug at issue is not a principled basis for treating similar statistics differently under Rule 403.

Second, Tran argues that the statistical evidence in Weinstock contained a “far greater volume of

data . . . covering all aspects of the defendant’s prescription practice.” But the government

offered complementary evidence—recorded phone calls as well as testimony from patients,

Highland Park pharmacists, and expert witnesses—showing that many of Tran’s prescriptions

conformed to Patel’s financial interests rather than her patients’ medical needs. To the extent she

argues that the government’s focus on her Medicare patients presented the jury with a distorted

view of her practice, she was free to counter with evidence of her non-Medicare prescriptions

and to point out the methodological shortcomings of the government’s statistics on cross-

examination.

Tran also compares her case to two inapposite, extra-circuit decisions. See United States

v. McKay, 715 F.3d 807 (10th Cir. 2013); United States v. Jones, 570 F.2d 765 (8th Cir. 1978).

In McKay, the Tenth Circuit opined that charts showing how the defendant prescribed more

hydrocodone and oxycodone than any other Utah physician would have been inadmissible under

Rule 403 had the defendant not opened the door at trial. 715 F.3d at 841–42. But the McKay

court specifically contrasted the use of statistical evidence to prove distribution charges with its

use to prove a complex healthcare-fraud conspiracy in which doctors prescribed unnecessary or

excessive amounts of drugs. Id. at 841 (citing United States v. Merrill, 513 F.3d 1293, 1303

(11th Cir. 2008)). Thus, the Tenth Circuit distinguished McKay from the circumstances here: a

complex healthcare-fraud conspiracy in which Tran prescribed unnecessary hydrocodone in

combination with an excessive quantity of non-controlled drugs.

-6-
Case No. 14-1092
United States v. Tran

Jones is also distinguishable. In that case, the Eighth Circuit reversed a trial court’s

admission of evidence showing that the defendant authored 478 prescriptions for Schedule II

drugs over a twenty-two-month period. Jones, 570 F.2d at 767. The court noted that the

government introduced no evidence showing that the defendant issued any of these prescriptions

without a medical reason. Id. at 769. But here, as Tran concedes, the government produced

evidence that she wrote at least some prescriptions without a medical rationale. Neither these

decisions nor Tran’s publicity argument persuade us that the district court abused its

considerable discretion under Rule 403.

B. Expert Testimony

Tran next challenges the expert testimony of pharmacist Dr. William Drake, arguing that

it exceeded the scope of his expertise. See Fed. R. Evid. 702. The government called Drake to

testify about a pharmacist’s ethical and legal obligation to question or refuse prescriptions that

appear to be outside the regular course of a physician’s practice. In light of his training and

experience filling the prescriptions of podiatrists, Drake testified that he would have questioned

or refused Tran’s prescriptions because they raised ethical concerns. Tran objected to Drake’s

testimony on the ground that his experience as a pharmacist did not qualify him to opine on the

medical necessity of her prescriptions.2

On appeal, Tran stresses that “[w]hile Drake purported to offer his opinions from the

perspective of a pharmacist . . . , the import of [his] opinions was a judgment on the medical

2
Tran also insinuates that Drake misrepresented his medical training. But the statements
she characterizes as misinformation were Drake’s reasonable answers to questions asked by trial
counsel on cross-examination. (See R. 1161, Trial Tr. at 129–30, 137–40, ID 13120–21, 13128–
31.) Drake explained that he considered many aspects of his pharmacological experience to fall
within the term “medical training” and listed examples. He readily conceded that he lacked a
physician’s training.
-7-
Case No. 14-1092
United States v. Tran

decisionmaking . . . which underlay the issuance of [Tran’s] prescriptions.” She insists that

Drake was not a physician and, therefore, could not assess whether her prescriptions came within

the scope of her practice. But Tran cites no relevant legal authority in support of this argument

and it fails to persuade us that admission of this testimony amounted to an abuse of discretion.

Tran’s trial counsel conceded that “any opinion [Drake] gives . . . indicating that there

should be red flags raised[ by Tran’s prescriptions] is perfectly appropriate.” It logically follows

that Drake also possessed the requisite expertise to explain why Tran’s prescriptions raised “red

flags.” Drake’s ability to identify suspicious prescriptions flows from the rational application of

his training as a pharmacist. See Fed. R. Evid. 702(c)–(d) (requiring experts to “reliably appl[y]”

“reliable principles and methods”). Drake’s training qualified him both to remark on the

apparent invalidity of Tran’s prescriptions and to explain the basis for his assessment—i.e., that

the type and volume of Tran’s prescriptions appeared to fall outside the scope of a podiatrist’s

practice. Cf. United States v. Smith, 573 F.3d 639, 654 (8th Cir. 2009) (holding that a

pharmacist’s “working knowledge of what constitutes a valid prescription” could not “be

divorced from having an awareness as to the quantity and quality of patient information a doctor

must have in order to prescribe a particular drug”). We perceive no abuse of discretion in

admitting Drake’s testimony.

C. Tran’s Sentence

Finally, Tran challenges the calculation of her guidelines range. She contends that the

government failed to prove several facts underpinning the district court’s drug-quantity and

amount-of-loss findings. Specifically, Tran disputes that every unit of hydrocodone she

prescribed during the conspiracy was unlawfully distributed and that every prescription filled at

-8-
Case No. 14-1092
United States v. Tran

Highland Park constituted loss to the defrauded insurers. We conclude that any error by the

district court proved harmless.

We review drug-quantity findings for clear error. United States v. Swanberg, 370 F.3d

622, 625 (6th Cir. 2004); see also U.S.S.G. § 2D1.1, cmt. n.5. The government bears the burden

of proving the quantity of illegally dispensed drugs attributable to the defendant’s conduct by a

preponderance of the evidence. Swanberg, 370 F.3d at 625. Courts may draw all reasonable

inferences from the evidence—or lack thereof—presented. See United States v. Huffman, 529 F.

App’x 426, 431 (6th Cir.) (“[I]t was not inappropriate for the court to consider the lack of

evidence that some of the prescriptions were legitimate.”), cert. denied, 134 S. Ct. 320 (2013);

United States v. Leal, 75 F.3d 219, 229 (6th Cir. 1996) (“[T]he district court could have drawn

certain inferences, [even though they were] not explicitly set forth during sentencing.”),

abrogated on other grounds by United States v. Ellerbee, 73 F.3d 105, 107 (6th Cir. 1996); cf.

Swanberg, 370 F.3d at 625 (“Although Swanberg objected to the Presentence Report’s

calculation of the drug quantity attributable to him, he produced no contradictory evidence at the

hearing.”).

Although Tran acknowledges prescribing all 213,548 units of hydrocodone found by the

district court, she maintains that “the record does not even come close to demonstrating an

across-the-board lack of medical necessity” for them. But to show prejudicial error in her

guidelines range, Tran would need to prove that more than half—or 113,548 units—of the

district court’s drug-quantity finding is erroneous. See U.S.S.G. § 2D1.1(c)(7) (2013) (dictating

a sixteen-point base offense level for an amount between 100,000 and 400,000 units); Leal, 75

F.3d at 229 (upholding drug-quantity finding where the defendant needed to show clear error

with respect to over a thousand kilograms of controlled narcotics and the record showed, at best,

-9-
Case No. 14-1092
United States v. Tran

de minimis error). She directs us to nothing in the record that supports such a dramatic

reduction.

The government’s statistical evidence showed that Tran prescribed twice as much

hydrocodone to her Medicare patients as the second-ranking podiatrist in Michigan, her partner

Dr. Meghan Lee. Lee refused to participate in Patel’s conspiracy. Although she served the same

community of patients, her hydrocodone prescriptions totaled half of Tran’s, and these numbers

include the prescriptions Lee renewed when covering Tran’s patients. This evidence suggests

that the difference in hydrocodone prescriptions between the doctors resulted from Tran’s

participation in the conspiracy.

Furthermore, the government proved that the bulk of Tran’s prescriptions followed a

specific pattern: a controlled pain killer—usually hydrocodone—combined with a bundle of

eight to twelve expensive, non-controlled drugs. Drake testified that these prescription

combinations would raise “red flags,” and the Highland Park pharmacists admitted to shorting

the non-controlled drugs in prescriptions that fit this pattern. Through reasonable extrapolation,

then, the conspiracy motivated at least half of all Tran’s hydrocodone prescriptions. See United

States v. Rodriguez-Iznaga, 575 F. App’x 583, 588 (6th Cir. 2014) (allowing for the

extrapolation of evidence on reasoned, factually supported grounds), cert. denied, 135 S. Ct.

1167 (2015). Thus, the district court committed no clear error in finding at least 106,774 units

unlawfully dispensed. And because any figure over 100,000 units returns the same adjusted

offense level, we need not consider whether the district court erred in counting the rest.

The same is true for the district court’s loss calculation of $4,475,193.40. So long as the

record supports a finding of at least $1 million, Tran’s combined offense level remains

unchanged. See U.S.S.G. §§ 2B1.1(b)(1)(I), 3D1.4(a). District courts need provide only a

- 10 -
Case No. 14-1092
United States v. Tran

reasonable estimate of loss, which we review for clear error. Id. § 2B1.1, cmt. n.3(C); United

States v. Poulsen, 655 F.3d 492, 512 (6th Cir. 2011). Accordingly, Tran “must carry the burden

of demonstrating that the court’s evaluation of the loss was not only inexact but outside the

universe of acceptable computations.” United States v. Martinez, 588 F.3d 301, 326 (6th Cir.

2009) (quoting United States v. Raithatha, 385 F.3d 1013, 1024 (6th Cir. 2004)) (internal

quotation marks omitted).

As the government pressed at oral argument, the district court reasonably counted as loss

at least eighty-five percent of Tran’s Medicare prescriptions—a measure worth $1,711,499.65.

As discussed above, each prescription in this subset included a controlled substance like

hydrocodone, and therefore fit the fraudulent pattern identified by the government’s witnesses.

As further support for Tran’s offense level, her patients cost Medicare $1,069,644.42 more than

Lee’s patients in prescription-drug costs from May 2010 to August 2011. Once again, the cost of

Tran’s prescriptions surpassed all other podiatrists in Michigan during this period, with Lee

coming in a distant second. The documented period also accounts for less than half the time

Tran participated in the conspiracy. Combined with Tran’s apparent disregard for drug-seeking

behavior in her patients receiving controlled substances, this evidence supports the finding that

she wrote at least eighty-five percent of her Medicare prescriptions in furtherance of the

conspiracy. And Tran offers nothing to show the medical necessity of any drug included in these

prescriptions. See U.S.S.G. § 2B1.1, cmt. n.3(E) (allowing defendants to reduce the amount of

loss by the fair market value of services actually rendered); id. § 2B1.1, cmt. n.3 (F)(viii) (stating

that fraudulent healthcare bills constitute prima facie evidence of intended loss); United States v.

Washington, 715 F.3d 975, 984–85 (6th Cir. 2013) (holding that defendants bear the burden of

proving what part of a fraudulent bill represents services rendered).

- 11 -
Case No. 14-1092
United States v. Tran

With no clear error evident in at least $1,711,499.65 of the district court’s loss finding,

Tran would have received at least a sixteen-point enhancement on the fraud count, leaving her

total offense level and guidelines range unaltered. Thus, any error in calculating the remainder

was harmless.

III.

We AFFIRM the district court’s judgment.

- 12 -