United States v. Peter E. Clay

Related Cases

    Case: 14-12373 Date Filed: 08/11/2016 Page: 1 of 124 [PUBLISH] IN THE UNITED STATES COURT OF APPEALS FOR THE ELEVENTH CIRCUIT ________________________ No. 14-12373 ________________________ D.C. Docket No. 8:11-cr-00115-JSM-MAP-5 UNITED STATES OF AMERICA, Plaintiff-Appellee, versus PETER E. CLAY, TODD S. FAHRA, PAUL L. BEHRENS, WILLIAM L. KALE, Defendants-Appellants. ________________________ Appeals from the United States District Court for the Middle District of Florida ________________________ (August 11, 2016) Before TJOFLAT and HULL, Circuit Judges, and HALL, * District Judge. * Honorable J. Randal Hall, United States District Judge for the Southern District of Georgia, sitting by designation. Case: 14-12373 Date Filed: 08/11/2016 Page: 2 of 124 HULL, Circuit Judge: In this Medicaid fraud case, defendants Todd Farha, Paul Behrens, William Kale, and Peter Clay appeal their convictions on multiple grounds, including insufficient evidence, evidentiary errors, and improper jury instructions. At the time of the fraud, the defendants were all high-level executives of WellCare Health Plans, Inc. (“WellCare”) or one of its two Florida subsidiaries. Those subsidiaries were Wellcare of Florida, Inc. doing business as Staywell Health Plan of Florida (“Staywell”) and HealthEase of Florida, Inc. (“HealthEase”). At trial, the government proved that together the defendants participated in a fraudulent scheme to file false Medicaid expense reports that misrepresented and overstated the amounts Staywell and HealthEase spent on medical services for Medicaid patients, specifically outpatient behavioral health care services. By overstating these expenses, the defendants helped Staywell and HealthEase retain millions of dollars in tax-subsidized Medicaid funds that they should have refunded to the Florida Agency for Health Care Administration (“AHCA”). This, in turn, inflated the profits of Staywell, HealthEase, and WellCare and earned the defendants financial rewards. The jury found Farha, Behrens, and Kale guilty on two counts of substantive health care fraud and found Behrens and Clay guilty on two counts of making false representations or statements. 2 Case: 14-12373 Date Filed: 08/11/2016 Page: 3 of 124 After reviewing the extensive trial record and with the benefit of oral argument, we affirm the defendants’ convictions. I. PROCEDURAL HISTORY A. Indictment On March 2, 2011, a federal grand jury in the Middle District of Florida returned an 11-count indictment against defendants Farha, Behrens, Kale, and Clay. The defendants were executives at WellCare, a publicly-held corporation headquartered in Tampa, Florida. Todd Farha was CEO and President of WellCare and one of its directors. Farha assumed leadership at WellCare in July 2002. Paul Behrens was CFO. Behrens joined WellCare in September 2003. Both Farha and Behrens held similar positions with Staywell and HealthEase, WellCare’s two subsidiaries. William Kale was Vice President of Clinical Services at WellCare. Kale joined WellCare in the fall of 2002. Peter Clay joined WellCare in April 2005 as Vice President of Medical Economics and reported to Behrens. Count 1 of the indictment charged the defendants with conspiracy to defraud the United States, to make false statements relating to health care matters, and to commit Medicaid health care fraud from 2003 through 2007, in violation of 18 U.S.C. § 371.1 Counts 2 through 5 charged the defendants with making false statements in Medicaid health care expense reports submitted to state officials, in 1 The indictment also charged Thaddeus Bereday, WellCare’s general counsel and a Senior Vice President. Bereday’s case was severed and is not at issue here. 3 Case: 14-12373 Date Filed: 08/11/2016 Page: 4 of 124 violation of 18 U.S.C. §§ 1035 and 2. Counts 2 and 3 covered the calendar year (“CY”) 2005 reports, and Counts 4 and 5 covered the CY 2006 reports. 2 Counts 6 through 9 charged the defendants with Medicaid health care fraud, in violation of 18 U.S.C. §§ 1347 and 2. Counts 6 and 7 covered CY 2005, and Counts 8 and 9 covered CY 2006. Counts 10 and 11 charged Clay with making false statements to federal agents in 2007, in violation of 18 U.S.C. § 1001. B. Jury Verdict After a trial lasting almost three months, the jury returned a mixed verdict. It was unable to reach a verdict as to any defendant on Count 1, the conspiracy charge. The jury acquitted the defendants of Counts 2 and 3, involving the CY 2005 expense reports. As to Counts 4 and 5, involving the CY 2006 expense reports, the jury convicted Behrens, acquitted Farha, and was unable to reach a verdict as to Clay and Kale. As to Counts 6 and 7, involving the health care fraud in CY 2005, the jury acquitted Farha and Kale, and was unable to reach a verdict as to Behrens and Clay. 2 Since WellCare made submissions for both Staywell and HealthEase each calendar year, the defendants were charged separately for submissions made on behalf of each company. For Counts 2 through 9, the even counts pertained to Staywell and the odd counts pertained to HealthEase. For purposes of our analysis, these distinctions do not matter since each defendant’s conduct generally related to both companies. We consider each of the pairings—Counts 2 and 3, Counts 4 and 5, Counts 6 and 7, and Counts 8 and 9—together. 4 Case: 14-12373 Date Filed: 08/11/2016 Page: 5 of 124 As to Counts 8 and 9, involving the health care fraud in CY 2006, the jury convicted Behrens, Farha, and Kale, but was unable to reach a verdict as to Clay. As to Counts 10 and 11, the jury convicted Clay of making false statements to federal agents in 2007. In sum, Behrens was convicted of Counts 4 and 5, making false statements in the Medicaid CY 2006 reports, in violation of 18 U.S.C. §§ 1035 and 2; Behrens, Farha, and Kale were convicted of Counts 8 and 9, Medicaid health care fraud in CY 2006, in violation of 18 U.S.C. §§ 1347 and 2; and Clay was convicted of Counts 10 and 11, making false statements to federal agents in 2007, in violation of 18 U.S.C. § 1001. After trial, the defendants filed renewed Rule 29(c) motions for judgment of acquittal, which the district court denied. The district court eventually dismissed all counts on which the jury was unable to reach a verdict. C. Sentences The district court sentenced the defendants well below their advisory guidelines ranges. The district court sentenced: (1) Farha to three years’ imprisonment on Counts 8 and 9 (to run concurrently), two years’ supervised release, and a $50,000 fine; (2) Behrens to two years’ imprisonment on Counts 4, 5, 8, and 9 (to run concurrently) and two years’ supervised release; (3) Kale to a prison term of one year and one day on Counts 8 and 9 (to run concurrently) and 5 Case: 14-12373 Date Filed: 08/11/2016 Page: 6 of 124 two years’ supervised release; and (4) Clay to five years’ probation on Counts 10 and 11 (to run concurrently), 200 hours of community service, and a $10,000 fine. Farha and Clay paid their fines. The defendants appeal their convictions, primarily challenging the sufficiency of the evidence. We thus recount the trial evidence in great detail. II. MEDICAID PROGRAM IN FLORIDA The Medicaid program is a cooperative federal and state health care benefit program, which assists states in paying for and providing medical services to qualifying, often disabled or low-income, individuals and families. While the program is jointly run, the federal government provides most of the funding. As part of the U.S. Department of Health and Human Services, the Centers for Medicare & Medicaid Services (“CMS”) authorizes and administers the states’ Medicaid programs. The states must regularly report to CMS regarding their expenses and operations. If a state Medicaid program does not expend all of its federal money in a given reporting cycle, the state must refund that money to the federal government. In Florida, AHCA administers the state Medicaid program. AHCA contracts with a variety of private health care companies, known as managed care organizations or health maintenance organizations, such as Staywell and 6 Case: 14-12373 Date Filed: 08/11/2016 Page: 7 of 124 HealthEase, to pay health care providers for the care delivered to Medicaid patients. For our purposes, we refer to these entities as HMOs. Medicaid and, in turn, AHCA cover medical and behavioral health care services. This case involves expense reports for only two types of outpatient behavioral health care services: (1) Community Mental Health (“CMH”) services, and (2) Targeted Case Management (“TCM”) services. We refer to them as “CMH/TCM” services. 3 A. AHCA Contracts Staywell and HealthEase operated under contracts with AHCA to cover medical and behavioral health care services for Medicaid enrollees. Staywell and HealthEase received a monthly premium from AHCA. AHCA calculated the premium, which is sometimes called a “capitation” payment, based on the number of Medicaid patients Staywell and HealthEase covered. For each covered member, AHCA paid a flat, capitated rate, known as a per-member-per-month or “PMPM” payment. This flat capitated rate was based on the estimated cost of providing a typical Medicaid patient’s needed health care services and did not vary based on Staywell’s and HealthEase’s actual costs for covered members. 3 CMH services are provided at a Community Mental Health Center and include certain medical, psychiatric, behavioral health therapy, community support and rehabilitation, therapeutic behavioral on-site day treatment, crisis intervention, and substance abuse services. TCM is intensive outpatient care provided by certified personnel trained to provide one- on-one life-coordination services, including home visits, to high-risk patients with severe emotional or mental conditions. 7 Case: 14-12373 Date Filed: 08/11/2016 Page: 8 of 124 This capitation system allowed AHCA to shift risk to Staywell and HealthEase. If Staywell and HealthEase on average spent more per enrolled Medicaid patient than the capitated rate, they would incur a loss. But if they spent less, they made a profit. In theory, AHCA was incentivizing Staywell and HealthEase to provide preventive care to decrease total health care costs. Staywell and HealthEase used different methods to provide behavioral health care services to patients. Staywell contracted directly with health care providers. Staywell reimbursed some providers on a fee-for-service basis but paid other providers a flat sub-capitated rate for each patient treated. HealthEase, on the other hand, subcontracted with CompCare, an independent behavioral health organization (“BHO”) with a network of providers. HealthEase paid CompCare a sub-capitated rate per enrolled patient, and, in turn, CompCare subcontracted with its network’s providers to treat HealthEase’s Medicaid patients. A sub-capitation arrangement with a subcontractor mirrors a capitation arrangement, but the rate is lower and the suite of covered services is generally more limited. As of July 1, 2002, AHCA’s contracts started requiring coverage for the two types of outpatient behavioral health care at issue here, CMH/TCM services. AHCA identified what particular services would qualify as CMH/TCM services in two coverage and limitations handbooks. 8 Case: 14-12373 Date Filed: 08/11/2016 Page: 9 of 124 In exchange for this new coverage obligation, AHCA increased the capitated rate for behavioral health care. AHCA piloted the CMH/TCM program in a limited geographic area (called Areas 1 and 6) that included Pensacola and Tampa. For reporting purposes, AHCA notified Staywell and HealthEase each year what portion of the capitation payment was intended to cover CMH/TCM services. B. Florida’s 80/20 Rule CMH/TCM services were a very profitable part of Staywell’s and HealthEase’s business. But those profits were threatened when Florida enacted restrictions on companies that received Medicaid money. Effective June 7, 2002, Florida amended its Medicaid statute as to “comprehensive behavioral health care services.” This amendment, which created the “80/20 rule,” was intended to ensure that most Medicaid money was spent on patients’ medical treatment rather than yielding high profits for HMOs. 2002 Fla. Laws 4662, 4693-94. To achieve this goal, the 80/20 rule required AHCA to include in its contracts a requirement that an HMO spend at least 80% of its capitation payment on providing behavioral health care services. If an HMO spent less than 80% of the premium on behavioral health care services, the HMO was required to refund the difference to AHCA. An HMO could retain no more than 20% of the premium for administrative costs, overhead, and profit. The 80/20 law read as follows: 9 Case: 14-12373 Date Filed: 08/11/2016 Page: 10 of 124 To ensure unimpaired access to behavioral health care services by Medicaid recipients, all contracts issued pursuant to this paragraph shall require 80 percent of the capitation paid to the managed care plan, including health maintenance organizations, to be expended for the provision of behavioral health care services. In the event the managed care plan expends less than 80 percent of the capitation paid pursuant to this paragraph for the provision of behavioral health care services, the difference shall be returned to the agency. Fla. Stat. § 409.912(4)(b) (2006). Upon the amendment’s enactment, AHCA’s contracts with Staywell and HealthEase imposed the 80/20 rule on only premium money for outpatient behavioral health care services, specifically CMH/TCM services. AHCA required Staywell and HealthEase annually to submit expense reports certifying that 80% of the AHCA premium was spent on CMH/TCM services. To facilitate and standardize expense reporting, AHCA annually provided Staywell and HealthEase with a spreadsheet template (the “Worksheet”). The Worksheet was designed to calculate the portion of the premium Staywell or HealthEase spent on CMH/TCM treatment that year and the amount of any refund due to AHCA. To illustrate the expense-reporting process, we discuss Staywell’s Worksheet for CY 2006. The Worksheet had five line items: (1) AHCA’s CY 2006 capitation payment to Staywell for CMH/TCM services; (2) the total amount Staywell spent on CMH/TCM services in CY 2006; (3) the ratio of line 2 to line 1, expressed as a percentage; (4) the difference between line 3 and the 80% 10 Case: 14-12373 Date Filed: 08/11/2016 Page: 11 of 124 minimum ratio; and, (5) if line 3 was less than 80%, the refund Staywell owed AHCA to reach the 80% minimum. The Worksheet for CY 2006 appears below: The Worksheet referenced the 80/20 rule and instructed Staywell that the purpose of the Worksheet was to determine whether it had spent at least 80% of its premium on “only” CMH/TCM services, stating: Pursuant to Section 409.912(4)(b), F.S., managed care entities that provide behavioral health services must expend at least eighty (80) percent of the capitation paid by the Agency on those services, defined as community mental health and targeted case management services only. If less than eighty (80) percent of the capitation is expended on these services, the entity shall return the difference to the Agency. (emphasis added). The Worksheet required Staywell’s CEO or President to certify the accuracy of Staywell’s reported expenses. When AHCA sent the Worksheet to Staywell or HealthEase, AHCA had already filled in line 1, identifying how much premium money AHCA had paid 11 Case: 14-12373 Date Filed: 08/11/2016 Page: 12 of 124 them for CMH/TCM services. All Staywell and HealthEase had to do was fill in their actual expenses on line 2. The rest of the calculations automatically flowed from those two numbers. This case concerns the defendants’ fraudulent reporting of false and inflated expenses on line 2 to keep Staywell and HealthEase from having to pay larger refunds. In July 2002, shortly after the 80/20 rule took effect, Farha joined WellCare as CEO. Later that fall, Farha’s team acquired Staywell and HealthEase. During Farha’s tenure, Farha signed several amendments to the Staywell and HealthEase contracts with AHCA, wherein Farha as CEO repeatedly agreed to the contracts’ underlying terms. C. Profit and Refund Studies In the spring of 2003, Farha asked WellCare actuary Todd Whitney to analyze Staywell’s and HealthEase’s profitability as to their Medicaid components. On May 7, 2003, Whitney emailed Farha a spreadsheet titled “FL Medicaid Projected Behavioral Health Profit.” The spreadsheet tracked what Whitney called the “contribution margin,” that is, premium revenue for behavioral health minus Medicaid claim costs. Whitney’s calculations revealed how much of the premium payment Staywell and HealthEase kept for administrative costs, overhead, and profit after paying medical claims. 12 Case: 14-12373 Date Filed: 08/11/2016 Page: 13 of 124 As to Staywell, Whitney’s calculations showed that, after paying all CMH/TCM claims, in some areas of Florida Staywell was keeping approximately 70% of its premium money for administration, overhead, and profit (much more than the 20% that the 80/20 rule allowed). For CMH/TCM claims, Staywell’s most profitable area was Area 6, in which Staywell received $15.00 per-member- per-month (“PMPM”) but paid on average only $4.69 PMPM. In Area 6, Staywell paid only 31.3% of its premium on CMH/TCM claims and retained the remaining 68.7% for administration, overhead, and profit. Given Staywell’s total membership in Area 6, Staywell’s annual contribution margin in Area 6 was $5,925,691, almost double its margin in all other areas of Florida combined. HealthEase had similar results. Staywell’s and HealthEase’s large contribution margins for Areas 1 and 6 were due to the much higher capitated rates of $15.00 PMPM that AHCA paid for Areas 1 and 6, as opposed to $4.00 PMPM for all other areas. The additional $11.00 PMPM more than made up for the marginal increase in claim costs in Areas 1 and 6, the areas where AHCA required coverage of CMH/TCM services. WellCare executives quickly recognized the implications of Florida’s new 80/20 rule. As early as February 2003, Kale circulated an email expressing concern about WellCare’s “potential exposure regarding the new requirement that Medicaid HMO’s must expend 80% of the capitation for [CMH/TCM] services.” 13 Case: 14-12373 Date Filed: 08/11/2016 Page: 14 of 124 Kale projected a potential refund to AHCA of almost $6.5 million (enough to dramatically reduce WellCare’s large behavioral health care profits). Thereafter, Whitney evaluated various refund scenarios for Staywell and HealthEase in Areas 1 and 6. The scenarios considered different definitions of CMH/TCM expenses. From July 2002 through September 2003, based on a strict definition of CMH/TCM expenses, Staywell had spent just 23% of its premium on CMH/TCM expenses and would have to pay back as much as $6,289,863. In the best case scenario, based on a looser definition of CMH/TCM expenses, Staywell had spent just 36% on CMH/TCM expenses and would have to pay back at least $4,803,645, or $400,000 per month. D. Creating New Subsidiary In light of the size of the potential refunds, WellCare began setting up a scheme to evade the 80/20 rule and keep its large profits. Under the scheme: (1) WellCare would create a new wholly-owned subsidiary; (2) Staywell and HealthEase would transfer their provider contracts to the new subsidiary; (3) Staywell and HealthEase would each pay 85% of their premium received for CMH/TCM services to WellCare’s new subsidiary; and (4) the new subsidiary would continue to pay the much smaller portion of the premium for CMH/TCM services. This structure would enable Staywell and HealthEase to report expenses in excess of 80%, while the new subsidiary would continue to pay only 45% or less 14 Case: 14-12373 Date Filed: 08/11/2016 Page: 15 of 124 directly to providers. Under the scheme, WellCare would preserve its large profit margins in these two types of behavioral health care services in spite of the new 80/20 rule. The defendants began planning for the new subsidiary at least as early as mid-2003. On July 16, 2003, Farha emailed Kale stating, “[W]e really need to think about how to setup a BH [behavioral health] subsidiary, that will be capped at 80% of premium.” Kale responded, “OK Todd…” By the fall of 2003, Farha grew impatient with the slow progress of implementation. On September 17, 2003, Farha sent an email to Kale with a subject line reading, “Status of BH Subsidiary / Need update.” Kale responded that the incorporation documents for the new subsidiary, “WellCare Behavioral Health, Inc. (WCBH),” were near completion and that outside counsel would begin drafting contracts for Staywell and HealthEase to subcontract with WCBH. Kale also explained that “a subsidiary corp is necessary for our Areas 1 & 6 programs” but that this “would change if the State would somehow repeal the 80% . . . requirement . . . .” Farha imposed a deadline: “Bill, Given the stakes involved (potentially 400k/Month of giveback), the pace of this project is not acceptable. We must execute these intercompany contracts asap, and get this subsidiary operating by 10/1. Why would we delay and increase the amount of our potential giveback? We must finalize this.” (emphasis added). Farha sent an even testier 15 Case: 14-12373 Date Filed: 08/11/2016 Page: 16 of 124 follow-up message to general counsel Thad Bereday: “This Goddamn thing is costing us 400K/Month. OUTSOURCE: Get it done, GT/ OTher/ Spend $$. I don’t care. This is absolutely stupid.” On September 22, 2003, Kale wrote Farha: “As we agreed, setting up the corporation is easy; it is the questions that follow (and probably many more not included in this work plan) that will determine if we create a viable organization if we were to be audited by AHCA.” 4 In September 2003, WCBH was finally incorporated. Farha was WCBH’s president, CEO, and director-chairman. Behrens later became CFO and a director. Kale became Vice President of Clinical Operations. Like Staywell and HealthEase, WCBH did not provide any Medicaid-reimbursable health care services.5 Lest there be any doubt, a WellCare slide titled “Fund Allocation Model” painted a clear picture of how WellCare was creating and using this new subsidiary to evade the 80/20 rule: 4 The “work plan” to which Kale referred appears to be a WellCare document titled “Behavioral Health Subsidiary Corporation Work Plan.” It included such action items as (a) “Develop Business Justification for all lines of business”; (b) “Justify why HMOs would pay 80% premium”; and (c) “Separate sub-lease for WCBH: Advantages in AHCA audit of distinctly standalone space at WellCare.” 5 At one point, HealthEase owned and operated a clinic in the Pensacola area. This exception aside though, Staywell and HealthEase were HMOs and did not provide health care services to patients and certainly not CMH/TCM services. 16 Case: 14-12373 Date Filed: 08/11/2016 Page: 17 of 124 WellCare’s slide shows that WellCare’s Staywell and HealthEase would: (1) receive the full premium from AHCA; (2) keep 15% for administration and overhead; and (3) pay 85% to WCBH. In turn, WCBH would pay only 45% of the whole for “direct behavioral health care services” and would keep 40% for administration, overhead, and profit. 6 Under this fund allocation scheme, WellCare entities retained 55% of the behavioral health care premium for administration, overhead, and profit, well over the 20% the 80/20 rule permitted. 6 As the slide shows, out of that 40%, WCBH kept 24% and paid 16% to CHMI, another internal WellCare subsidiary, for management. 17 Case: 14-12373 Date Filed: 08/11/2016 Page: 18 of 124 A jury could reasonably infer that Whitney’s $400,000-per-month refund projection spurred the creation of the new subsidiary. A company email explained that the 85% rate paid to the new WCBH subsidiary was “based on the historical premiums received by” Staywell and HealthEase from AHCA and “based on a conceptual pass through of 85%” of the total premium received from AHCA. (emphasis added). As WellCare financial analyst Greg West testified, the 85% pass-through figure was “[s]o [WellCare] wouldn’t pay anything back on the 80/20 payback.” Staywell and HealthEase each used a sub-capitated rate to pay WCBH. West testified he was told that the sub- capitated rate Staywell and HealthEase each paid WCBH was a “back-of-the- envelope calculation,” which to him meant the kind of “calculation you do in your head or on a piece of paper that you’re going to throw away; so you have no record of how it was calculated. And also that that would be round numbers, it wouldn’t be real-specific.” Another WellCare internal slide presentation framed WCBH as WellCare’s “[p]ro-active response to potential implications” of the “New Medicaid Mental Health Law in Florida” (the 80/20 rule). The slides listed as an action item that WellCare needed to “[p]repare [a] rationale for WCBH and answers to All AHCA inquiries, if any.” Staywell and HealthEase, by paying 85% of their behavioral 18 Case: 14-12373 Date Filed: 08/11/2016 Page: 19 of 124 health premium to WCBH, would pay at least twice as much as the market rates they would pay an independent, third-party BHO like CompCare. After WCBH was incorporated (and after the first round of 80/20 reporting discussed below), Farha instructed Kale to change WCBH’s name to Harmony Behavioral Healthcare—“and quickly.” Farha explained, “Let’s put some distance between BH [Harmony] and the WellCare name.” On August 26, 2004, WCBH changed its name from WellCare Behavioral Health, Inc. to Harmony Behavioral Health, Inc. (“Harmony”). III. 80/20 EXPENSE REPORTS Because the relevant limitations period precluded fraud charges relating to 2004 and earlier, the 2011 indictment charged the defendants with fraud only as to the CY 2005 and 2006 reports. We nevertheless consider the defendants’ conduct in submitting the CY 2002-04 reports because it shows their acquired knowledge and motive by the time they submitted the CY 2005 and 2006 reports. A. CY 2002 and 2003 Reports In 2004, Staywell and HealthEase each received a set of two Worksheets, one for expenditures from July 1 through December 31 of 2002 7 and one for all of 2003. The Worksheets showed on line 1 the amount of premium AHCA allocated 7 The letters accompanying the CY 2002 Worksheets explained that AHCA’s general counsel elected to impose the 80/20 requirements for only the second half of 2002, after the 80/20 law took effect, rather than for the full year, and AHCA had adjusted the premium figures in the Worksheets for CY 2002 accordingly. 19 Case: 14-12373 Date Filed: 08/11/2016 Page: 20 of 124 to CMH/TCM services. In a June 3, 2004 email, AHCA reminded Staywell and HealthEase that they were “required to expend at least 80 percent of the capitation paid on such services.” A cover letter reminded Staywell and HealthEase of their 80/20 obligations and explained how to fill out the Worksheets. The cover letters quoted the contract language relating to the 80/20 rule: By April 1 of each year, plans with members in Areas 1 and 6 shall provide a breakdown of expenditures related to the provision of behavioral health care, using the spreadsheet template provided by the agency. Pursuant to Section 409.912(3)(b), F.S., 80 percent of the capitation paid to the plan shall be expended for the provision of behavioral health care services. In the event the plan expends less than 80 percent of the capitation, the difference shall be returned to the agency. The letters explained that “[f]or reporting purposes, behavioral health care services are defined as those services the plan is required to provide, as listed in the Community Mental Health and Targeted Case Management Services Coverage and Limitations Handbooks.” To stress that AHCA wanted to know what the providers were paid, the letter added that “[a]s used above, expended means the total amount, in dollars, paid directly or indirectly to behavioral health providers for the provision of those required behavioral health care services.” The letters invited Staywell and HealthEase to contact AHCA if they had any questions. Upon receiving the CY 2002 and CY 2003 Worksheets, Pearl Blackburn, WellCare’s Director of Regulatory Affairs for Medicaid, filled out a “Regulatory Inquiry Routing Form” marked “Follow-up Required: Urgent” with topic 20 Case: 14-12373 Date Filed: 08/11/2016 Page: 21 of 124 “Behavioral Health Expenditures” and forwarded the Worksheets to several WellCare executives, including Farha, Behrens (identifying him as the “owner” of the 80/20 reporting project), Harmony executive Dave Smith, and general counsel Thad Bereday. On June 16, 2004, Bereday emailed Farha, Behrens, and others to inform them that their “team ha[d] been activated on the BH [behavioral health] expenditures reconciliation.” Bereday explained that “they [were] already busy calculating [their] BH expenditures to achieve the most favorable reporting possible to the state.” Bereday added, “I have also discussed this matter with Paul [Behrens] . . . . Paul will serve as the overall project lead.” The team responsible for calculating the 80/20 expenses was Medical Economics, a division of WellCare’s Finance Department, which Behrens oversaw. The team’s work largely fell to Smith, West, Kale, and another employee. Smith told West that Darrell Lettiere, a WellCare employee, had previously conducted a refund analysis and estimated that Staywell and HealthEase would collectively owe a $10.2 million refund. Smith told West that they had been “charged by Todd Farha to find a way not to pay back 10 million dollars.” They had to “find[] a way to make it zero.” West examined Lettiere’s refund analysis and discovered that it included a number of questionable 80/20 expenses. West noticed that Lettiere’s expense totals included not only payments to medical providers but also the amounts 21 Case: 14-12373 Date Filed: 08/11/2016 Page: 22 of 124 Staywell and HealthEase had paid to Harmony for the last two months of CY 2003. In response to West’s questions, Smith explained that WellCare had created Harmony as its own mental health company and Staywell and HealthEase had each paid Harmony 85% of the premium money they received from AHCA so “they didn’t have to pay it back.” Lettiere’s analysis still resulted in a $10 million projected refund because Harmony had existed for only a few months of CY 2003. To reduce the refund as close as possible to zero, as Farha requested, the team needed to include additional non-qualifying expenses. To reduce the refund, Kale told West to add in such non-qualifying items as: (1) a portion of all the pharmacy costs that correlated to the percentage of claims physicians submitted relating to behavioral health care; (2) both fee-for-service and capitation payments to primary-care physicians, including claims in which only a secondary diagnosis related to mental health (thus, for example, WellCare would include its payments for a claim involving a “broken arm” if the physician had included “depression” as a secondary diagnosis); and (3) claims either (a) paid to a mental health provider, (b) involving a mental health diagnosis, or (c) using a mental health procedure code, even though the CMH and TCM handbooks required all three elements for a claim to be considered a qualifying expense. West characterized these expenses as “gray areas” and “questionable items,” or in some instances “not even remotely close to behavioral health” expenses. Years later, 22 Case: 14-12373 Date Filed: 08/11/2016 Page: 23 of 124 Kale, during a secretly-recorded conversation, admitted: “Yeah, I did that analysis, I . . . remember this all too well.” Kale added, “We got very creative.” After including all of these non-qualifying items, the team managed to reduce Staywell and HealthEase’s collective total refund figure for CYs 2002 and 2003 to $6,147,700. On behalf of the team, Smith emailed Behrens and Bereday their final figures. Bereday then emailed Farha: After much back and forth, there is not going to be further change. Kale is already waivering [sic] in his support of this number, there was difficulty obtaining verifiable data that we felt could survive audit, and Paul [Behrens] feels we are currently being as aggressive as possible while still defensible. Smith is bringing you the certification now that you need to sign. Farha responded, “ok.” Staywell and HealthEase completed their CY 2002 and 2003 Worksheets consistent with the spreadsheet that Kale, West, and Smith produced. Staywell reported to AHCA that it spent $1,848,330 (41.1% of its premium for CMH/TCM) on qualifying services in CY 2002 and $4,519,744 (50.5% of its premium for CMH/TCM) on qualifying services in CY 2003. This resulted in Staywell paying a $1,746,965 refund for CY 2002 and a $2,634,626 refund for CY 2003. HealthEase reported to AHCA that it spent $1,663,077 (57.9% of its premium for CMH/TCM) on qualifying services in CY 2002 and $3,684,423 (61.2% of its premium for CMH/TCM) on qualifying services in CY 2003. This resulted in 23 Case: 14-12373 Date Filed: 08/11/2016 Page: 24 of 124 HealthEase paying a $636,433 refund for CY 2002 and a $1,129,676 refund for CY 2003. The entities collectively refunded $6,147,700 for CY 2002 and 2003. Farha signed off on the Worksheets affirming that “the expenditure information reported is true and correct to the best of [his] knowledge and belief.” At trial, West testified that the 80/20 expenses Staywell and HealthEase reported in their CY 2002 and CY 2003 Worksheets were “false number[s].” The government’s expert witness, Harvey Kelly, also testified the reported expenses were false. Kelly was a forensic accountant, CPA, and managing director at a financial consulting firm. Kelly reviewed and analyzed WellCare’s records, including its claims database. Based on his claims analysis, Kelly testified that the numbers WellCare reported were “not true and accurate,” bearing “no logical relationship . . . between monies paid to third-party providers for the provision of outpatient behavioral healthcare services.” While Staywell and HealthEase collectively reported an 80/20 expense total of $3,511,407 for CY 2002, their actual qualifying expenses totaled a mere $923,274, a difference of $2,588,133. The difference was even greater for CY 2003. Staywell and HealthEase reported an expense total of $8,204,167 for CY 2003, but their actual qualifying expenses totaled $3,350,656, a difference of $4,853,511. This means that in CY 2002 and CY 2003, Staywell and HealthEase over-reported their expenses by over $7 million and substantially underpaid their refunds. 24 Case: 14-12373 Date Filed: 08/11/2016 Page: 25 of 124 B. CY 2004 Reports AHCA renewed its contracts with Staywell and HealthEase for 2004. The new contract and the CY 2004 cover letter instructed: “For reporting purposes . . . ‘behavioral health services’ are defined as those services that the Plan is required to provide as listed in the Community Mental Health Services Coverage and Limitations handbook and the Targeted Case Management Coverage and Limitations handbook.” The new contract also instructed: “For reporting purposes . . . ‘expended’ means the total amount, in dollars, paid directly or indirectly to behavioral health providers solely for the provision of behavioral health services . . . not including administrative expenses or overhead of the plan.” (emphasis added). In January 2005, both Farha and Kale signed a WellCare “policy and procedure” document that mirrored the contract language. In February 2005, AHCA sent Staywell and HealthEase the CY 2004 Worksheets along with cover letters. The substance of the Worksheets and cover letters was essentially unchanged. As in CY 2002 and 2003, AHCA completed line 1 of the Worksheets, showing the CY 2004 premium amount paid to Staywell and HealthEase for CMH/TCM services. As she had during the previous reporting cycle, Pearl Blackburn routed the 80/20 reporting materials to Farha, Behrens (again, the “owner” of the project), and Kale. In response, on February 14, 2005, Farha emailed a group of people, 25 Case: 14-12373 Date Filed: 08/11/2016 Page: 26 of 124 including Behrens, Kale, Bereday, and Smith. Farha wrote: “Team, lets [sic] be sure we handle this one appropriately. Who is on point for this process?” Behrens replied: “Todd, I am on point for the completion of this required form. Specifically, Bill White is working with Medical Economics to assure timely and appropriate completion.” Smith and West were again tasked with compiling data for the reports. West testified that he had expected Staywell and HealthEase to report qualifying expenses totaling 85% of the premium each entity had received from AHCA. That was because, according to Smith, Staywell and HealthEase contracted with Harmony for the purpose of paying 85% to Harmony and avoiding a refund. For CY 2003, West had used the sub-capitated Harmony payments for the last two months of the year but otherwise counted an assortment of varied expense items for the reports. Because Harmony existed for all of CY 2004, and assuming Staywell and HealthEase had in fact paid Harmony 85% of their premium, West thought Staywell and HealthEase should refund nothing to AHCA. But Smith gave West different instructions. “The idea was to come up with a payback” after all. Smith told West to produce three preliminary refund scenarios based on different assumptions and generate total refunds of $0, $1 million, and $1.5 million. The idea was to refund at least some amount to AHCA (presumably to avoid an audit). Because reporting that Staywell and HealthEase 26 Case: 14-12373 Date Filed: 08/11/2016 Page: 27 of 124 had each paid Harmony 85% of their premium would result in no refund, West had to adjust downward from 85%. To manipulate the figures and create three refund scenarios, West relied on the fact that not all of Staywell’s and HealthEase’s payments to Harmony covered qualifying outpatient behavioral health care services. Staywell and HealthEase each paid Harmony a significant portion of premium for non-qualifying inpatient behavioral health care services, for which there was no AHCA reporting obligation. While the entities’ journal entries recorded the total amount Staywell and HealthEase each had paid Harmony, neither the records nor the entities’ contracts with one another distinguished between inpatient and outpatient payments. West therefore arbitrarily divided Staywell’s and HealthEase’s total respective payments into inpatient and outpatient portions, which West would then manipulate to create his refund scenarios. West created numerous spreadsheets titled “AHCA Behavioral Health (TCM and CMH) Payback Calculation.” Each spreadsheet identified a different portion of the CY 2004 premium for CMH/TCM as having been paid to Harmony: at 85%, Staywell and HealthEase would refund nothing; at 70%, they would collectively refund about $1 million; at 67%, they would collectively refund about $1.5 million. For each refund scenario, as West reduced the outpatient portion of Staywell’s and HealthEase’s sub-capitated payments to Harmony, he offset that reduction by 27 Case: 14-12373 Date Filed: 08/11/2016 Page: 28 of 124 increasing the inpatient portion. West never considered the actual amounts paid to health care providers for CMH/TCM services. West did not consult the Medicaid handbooks as he had the year before. The amounts Staywell and HealthEase actually paid (through Harmony) to health care providers for CMH/TCM services were not reflected in any of his three calculations. Smith later revised his instructions to West: the combined refund should total approximately $800,000, with Staywell and HealthEase each paying a portion, and the inpatient rates Staywell and HealthEase paid to Harmony should be the same. 8 These criteria had nothing to do with actual expenses for CMH/TCM services. West explained that Smith’s parameters required him to “back[] into” inpatient rates for both Staywell and HealthEase, increasing one HMO’s refund figure and decreasing the other’s until the inpatient rates were the same for both. West changed the numbers in his spreadsheets to comply with Smith’s instructions, thereby producing a fourth refund scenario. As Kelly, the forensic accountant, explained, West’s calculations focused not on determining qualifying expenses but on coming up with a desirable refund figure to AHCA. West discussed his work with Behrens, and Staywell’s and HealthEase’s final Worksheets were again based on West’s calculations. This time, Imtiaz 8 The total sub-capitated rate that Staywell and HealthEase each paid Harmony was not broken down into inpatient and outpatient rates in Harmony’s contracts. West thus manipulated the inpatient and outpatient rates to create these refund scenarios. 28 Case: 14-12373 Date Filed: 08/11/2016 Page: 29 of 124 Sattaur, then president of Staywell and HealthEase, signed instead of Farha. At trial, however, Sattaur testified that the work of WellCare’s Medical Economics team “would be approved by Mr. Paul Behrens, and the ultimate sign-off on the approval of whether [the Worksheets get] filed with the State would be by Mr. Todd Farha.” Staywell certified to AHCA that, in CY 2004, it spent $6,525,079 (72.1% of its premium for CMH/TCM) on qualifying services. Staywell therefore refunded $713,642 to AHCA. HealthEase certified that, in CY 2004, it spent $5,119,436 (79.0% of its premium for CMH/TCM) on qualifying services. HealthEase therefore refunded $65,707 to AHCA. The combined total expenses were $11,644,515 and the combined total refund was $779,349. West testified that the 80/20 expenses Staywell and HealthEase reported on their Worksheets were “false number[s].” Kelly, the forensic accountant, confirmed the falsity of Staywell’s and HealthEase’s reports. Based on an analysis of claims data, Kelly testified that Staywell’s and HealthEase’s actual CY 2004 qualifying expenses totaled only $3,522,000, a difference of $8,122,515. By over- reporting their expenses by over $8 million, Staywell and HealthEase substantially underpaid their refunds. WellCare’s own internal documents also confirmed the falsity of Staywell’s and HealthEase’s CY 2004 reports. Smith directed West to calculate for internal 29 Case: 14-12373 Date Filed: 08/11/2016 Page: 30 of 124 use Staywell’s and HealthEase’s “actual expenditures” in monies “actually being used for [CMH/TCM services].” West testified that he created a spreadsheet, partly with Clay’s input, which calculated Staywell’s and HealthEase’s CMH/TCM expenses according to the “strict definition” of qualifying expenses found in the CMH and TCM handbooks provided by AHCA. According to West’s spreadsheet, Staywell and HealthEase (through Harmony) had actually spent only $3,237,891.98 combined (19.9% of their premium) on CMH/TCM services in CY 2004, far below the $11,644,515 they reported to AHCA. West testified that, if claims for additional procedure codes provided by Kale were factored in, Staywell and HealthEase’s 80/20 expense percentage rose from 19.9% to 22.6%. Even if all of Harmony’s administrative costs were included, the percentage rose to only 51.1%. These percentages were still well short of the 72.1% and 79.0% expense percentages Staywell and HealthEase reported to AHCA in the Worksheets.9 Subsequently, Bereday shared with Farha a presentation that detailed Staywell’s and HealthEase’s reported expenses (72.1% and 79.0% respectively) and revealed what the entities’ “Medical Costs” were as defined by AHCA—that is, their actual qualifying expenses (19.9%, 22.6%, or 51.1%, per West’s analysis). Farha thus knew that Staywell and HealthEase had 9 As we discuss infra, there was a mathematical error in the premium figure AHCA listed on line 1 of the Worksheets for CY 2002 through 2004. Even if AHCA had listed the correct figure on line 1 of the Worksheets for CY 2004, Staywell’s and HealthEase’s qualifying expense percentages would still have been far below what they reported for CY 2004. Line 1 did not affect Staywell’s and HealthEase’s qualifying expenses. 30 Case: 14-12373 Date Filed: 08/11/2016 Page: 31 of 124 not reported their expenses for CMH/TCM services consistent with AHCA’s definition of qualifying expenses. C. CY 2005 Reports In mid-April 2006, AHCA sent Staywell and HealthEase the Worksheets for CY 2005 with instructional cover letters. Once again, the Worksheets listed “Targeted Case Management” and “Community Mental Health” as the only qualifying expenses on line 2. The Worksheets also defined “behavioral health services” as “community mental health and targeted case management services only.” As in prior years, AHCA completed line 1 of the Worksheets, showing how much premium Staywell and HealthEase received in CY 2005. While AHCA made minor wording changes to the Worksheet, AHCA revised the cover letter in some notable ways. The new cover letter now quoted language from the 80/20 law rather than from the AHCA contracts. Also, previous cover letters had instructed Staywell and HealthEase to use the CMH and TCM handbooks to determine which types of behavioral health care services qualified under the 80/20 rule. This time, the cover letter listed the only authorized procedure codes for eligible expenses, stating: The Agency has determined that for this purpose, “behavioral health care services” is defined as community mental health (procedure codes H0001HN; H0001HO . . . or T1023HF) and targeted case management (procedure codes T1017; T1017HA; or T1017HK). 31 Case: 14-12373 Date Filed: 08/11/2016 Page: 32 of 124 The AHCA contract in CY 2005 was the same one as CY 2004, and consequently still required Staywell and HealthEase to report only money paid to health care providers, not any administrative expenses or overhead. In mid-March 2006, before WellCare received the CY 2005 Worksheets, WellCare’s Medical Economics team started working on Staywell’s and HealthEase’s CY 2005 reports. West encountered several new hurdles. During CY 2005, AHCA had paid Staywell and HealthEase substantially more in capitation money for CMH/TCM services than previous years due to AHCA’s expanding the CMH/TCM program statewide. Although Staywell and HealthEase now covered CMH/TCM services for all of Florida (rather than just Areas 1 and 6), Staywell and HealthEase had not paid any of this new premium money to Harmony, which held the subcontracts with providers. In CY 2004, AHCA had allocated $15,529,829 as Staywell and HealthEase’s combined premium. But in CY 2005, West estimated that Staywell and HealthEase combined received $30,310,183, almost twice as much. When West calculated the prospective CY 2005 refunds using Staywell’s and HealthEase’s existing sub-capitation rates to Harmony and the same Harmony inpatient rates from CY 2004, West projected that Staywell and HealthEase would collectively owe AHCA an $11.9 million refund. West explained the problem to 32 Case: 14-12373 Date Filed: 08/11/2016 Page: 33 of 124 Clay and WellCare employee Bill White. White said, “[W]e should have changed our contract [with Harmony], and we didn’t.” West reported to Kale that if they wanted to refund nothing for CY 2005, they would have to reduce Harmony’s inpatient rate, which was $4.91 PMPM in CY 2004, to between $1.50 and $2.46 PMPM. Kale responded, “[T]his is good information.” Kale added, “If we wanted a small payback with an MLR below 80, we can attempt to justify a[n inpatient] number around 2.75 or 3.00. Thanks.” To avoid dramatically reducing the inpatient rate for both Staywell and HealthEase, the reporting team instead added (1) Staywell’s sub-capitation payments to Harmony of $7,337,954 for CMH/TCM services generally and (2) Harmony’s payments of $5,263,500 to health care providers in Areas 2-5 and 7-11, thereby manipulating Staywell’s total expense figure to be $12,601,454. For HealthEase, the team added (1) HealthEase’s sub-capitation payments to Harmony of $6,169,747 for CMH/TCM services generally and (2) Harmony’s payments of $5,122,816 to health care providers in Areas 2-5 and 7-11, thereby manipulating HealthEase’s total expense figure to be $11,292,563. At trial, Kelly, the forensic accountant, described this maneuver as a kind of “double counting.” Although Staywell and HealthEase had not actually paid Harmony any of the increased premium they had received for the CMH/TCM program expansion, Harmony nevertheless had covered CMH/TCM claims statewide. Kelly explained, “You 33 Case: 14-12373 Date Filed: 08/11/2016 Page: 34 of 124 can’t have it both ways. You can’t say . . . ‘I’m going to pay you for the capitation,’ and ‘oh, by the way, you know, if you pay any providers, I’ll tell the state I paid the providers too.’” With this method, West projected Staywell and HealthEase would owe a combined refund of $699,223, far less than the $11.9 million West had originally projected. West was optimistic about this calculation maneuver because the total projected refund amount was close to the previous year’s refund of almost $800,000 without dramatically affecting Harmony’s inpatient rate. In a group email that included Clay, West explained his work and wrote “I think we got it!” But not quite. West’s calculations were based on his estimate that Staywell and HealthEase had received a combined $30,310,183 in premium for CMH/TCM services for CY 2005. West estimated a $30,310,183 premium figure based on information from rate tables on AHCA’s website. On April 18, AHCA emailed Staywell and HealthEase the CY 2005 Worksheets. On line 1, AHCA allocated a $12,306,570 premium to Staywell and a $12,572,017 premium to HealthEase. The combined total premium of $24,878,587 was about $5.4 million less than West’s original $30,310,183 estimate. This $5.4 million difference between the actual premium figure on the Worksheets and West’s estimated premium figure came to be known as the 34 Case: 14-12373 Date Filed: 08/11/2016 Page: 35 of 124 “premium difference.” 10 Those both inside and outside of Medical Economics at WellCare did not know what to make of this premium difference between what AHCA said it had paid Staywell and HealthEase for outpatient behavioral health care, reflected on line 1, and what West estimated AHCA had paid. In the past, the premium figures on line 1 of the Worksheets had differed from West’s estimates by only a slight amount. Now, the difference substantially affected the refund calculation, resulting in neither Staywell nor HealthEase owing a refund. Despite their confusion, no one at WellCare called AHCA for clarification, even though the cover letters accompanying the Worksheets invited them to do so. From mid-April to mid-June 2006, the expense reporting team discussed what to make of this premium difference and whether it should factor into the expenses Staywell and HealthEase would report to AHCA. Of course, what Staywell and HealthEase actually spent on qualifying expenses was unrelated to the premium AHCA listed on line 1 of the Worksheets. Any change on line 1 would affect the HMOs’ refunds but not their qualifying expenses. Over the next several weeks, West and others considered a variety of refund scenarios. By mid-June, they found themselves up against the submission deadline 10 For the first few 80/20 reporting cycles, AHCA’s premium figures on line 1 of the 80/20 Worksheets were inaccurate due to an error in the rate tables on AHCA’s website. In prior years, the Worksheet premium figures included money for some inpatient behavioral health care services (non-80/20) in addition to the money Staywell and HealthEase received to cover outpatient behavioral health care, namely CMH/TCM services. As a result, the premium figures on the Worksheets were too high. For CY 2005 and years following, AHCA corrected that error and listed the correct premium figures on the 80/20 Worksheets. 35 Case: 14-12373 Date Filed: 08/11/2016 Page: 36 of 124 for Staywell’s and HealthEase’s Worksheets. Clay met with Farha and suggested that Staywell and HealthEase refund nothing for CY 2005. Farha disagreed, explaining to Clay, “No, we’re not going to do it like that. You have to pay the Gods something.” Clay passed Farha’s orders along to West: “Farha wants to pay back a million.”11 West was not sure how that request could be met. After rocking back and forth on his heels and glancing around for a few moments, Clay asked, “We have a premium difference, don’t we?” “Yeah,” West answered. Clay pressed, “Well, if you refunded that?” As discussed below, Clay instructed West to run the numbers using the premium difference calculation Clay had suggested. West testified that Clay then stared off into the distance and said to no one in particular, “[I] was told to find a million. [I] didn’t know how [I] could do it, and [I] did it.” Before encountering the premium difference, West had counted both (1) Staywell and HealthEase’s combined sub-capitation payments to Harmony, $13,507,701, and (2) Harmony’s fee-for-service payments to providers in Areas 2- 5 and 7-11, $10,386,316. Now, to reach Farha’s desired $1 million refund, Clay instructed West to subtract the premium difference from Harmony’s total fee-for- service payments in Areas 2-5 and 7-11. This calculation simply halved the fee- for-service costs that Staywell and HealthEase double counted and yielded the 11 West later testified that Behrens and Bereday also confirmed to West that Farha wanted to refund about one million dollars to AHCA for CY 2005. 36 Case: 14-12373 Date Filed: 08/11/2016 Page: 37 of 124 desired result, increasing the combined refund total for Staywell and HealthEase to about $1.4 million. As with other aspects of Staywell and HealthEase’s evolving expense reporting methodology, this premium difference calculation bore no relationship to what Staywell and HealthEase (through Harmony) had actually paid providers of CMH/TCM services or even to what Staywell and HealthEase had paid Harmony. Kelly, the forensic accountant, testified: “You have them including as components like the premium difference that has nothing to do with actual amounts expended or providing services.” On June 15, 2006, West, Behrens, and Clay reviewed the final numbers and then walked toward Bereday’s office. On the way, Behrens slipped into Farha’s office, and West overheard a discussion about “1.4.” Behrens rejoined the group and confirmed, “1.4 is okay.” As he looked over West’s spreadsheet, Bereday had questions. “I understand [Farha] wants to make a million dollar payback,” he said, but “I also see we’re refunding premium.” Bereday asked West about the premium difference and how confident West was about the premium estimates West had used in his refund calculations. West answered that the only way to be sure would be to call an AHCA financial analyst. “No,” Bereday told West, “[Y]ou’re not going to call . . . AHCA.” 37 Case: 14-12373 Date Filed: 08/11/2016 Page: 38 of 124 Because Sattaur was out that day, Bereday invited WellCare’s Jim Beermann into his office to certify the Worksheets. Bereday briefed Beermann on the Worksheets, explaining why WellCare had established Harmony and the components of the refund calculations, including the sub-capitation payments to Harmony, the double-counting calculation, and the premium difference calculation. West testified that after hearing all of this, Beermann looked “pretty uncomfortable,” and Beermann “backed himself up against the door, like he was trying to push himself out of the room.” Beermann suggested they wait for Sattaur to return so that he could certify the expense reports. But, according to West, Bereday, Behrens, and Clay immediately insisted, “No, no, it’s got to go today, you’re signing it.” Beermann relented and signed the certifications. Staywell certified to AHCA that it spent $9,587,573 or 77.9% of its premium for CMH/TCM on qualifying services in CY 2005 and refunded $257,683 to AHCA. HealthEase certified it spent $8,874,848 or 70.6% of its premium for CMH/TCM on qualifying services in CY 2005 and refunded $1,182,766 to AHCA. Combined, Staywell and HealthEase reported $18,462,421 in expenses and paid a $1,440,449 refund. At trial, West admitted that the expenses Staywell and HealthEase reported for CY 2005 had nothing to do with what they paid to providers for CMH/TCM services. Based on his analysis of claims data, Kelly, the forensic accountant, 38 Case: 14-12373 Date Filed: 08/11/2016 Page: 39 of 124 testified that, while Staywell and HealthEase together had reported $18,462,42 in CMH/TCM expenses for CY 2005, their actual qualifying expenses, based on what Harmony paid to health care providers, totaled $13,100,136, a difference of $5,362,285. By over-reporting their expenses by over $5 million, Staywell and HealthEase substantially underpaid their refunds. WellCare’s internal records also revealed Staywell’s and HealthEase’s CY 2005 reports were false and fraudulent. Starting with CY 2005, West’s internal spreadsheets included a calculation of Staywell’s and HealthEase’s qualifying expenses and corresponding refunds if they counted only the money Harmony paid to providers for CMH/TCM services. West’s spreadsheets revealed that their qualifying expenses were much less than they reported to AHCA. As both West and the Kelly explained at trial, West’s spreadsheets showed that Staywell and HealthEase combined (through Harmony) had paid to health care providers only $12,956,122 or 52.1% of their premium on CMH/TCM services, and that they should have refunded $6,946,748 to AHCA. It was no secret that Staywell and HealthEase truly owed $6,946,748. Only days before Beermann certified the Worksheets, Clay wrote Behrens, saying, “If we took AHCA payments and AHCA definitions of eligible care we would owe them $6.9 million.” Instead, due to Staywell’s and HealthEase’s false reporting, they refunded only $1,440,449 to AHCA. 39 Case: 14-12373 Date Filed: 08/11/2016 Page: 40 of 124 D. CY 2006 Reports We now turn to CY 2006, the reporting year for which Farha, Behrens, and Kale were convicted of health care fraud as to the false and fabricated expenses reported in the Worksheets, in violation of 18 U.S.C. § 1347, and Behrens was convicted of making false statements, in violation of 18 U.S.C. § 1035. This was the fourth year that Staywell and HealthEase reported to AHCA their qualifying expenses for CMH/TCM services. By this time, it was perfectly evident that AHCA wanted to know what Staywell and HealthEase were paying to health care providers. AHCA’s instructions were direct and unambiguous in three places: (1) the contract, (2) the Worksheets, and (3) the cover letters. For 2006, AHCA, Staywell, and HealthEase executed new contracts, which, as before, expressly instructed: “For reporting purposes . . . ‘expended’ means the total amount, in dollars, paid directly or indirectly to community behavioral health services providers solely for the provision of community behavioral health services, not including administrative expenses or overhead of the plan.” (emphasis added). AHCA’s requirement was clear: only money paid to health care providers for CMH/TCM services qualified. Staywell and HealthEase could not include administrative or overhead expenses. As in prior years, Farha signed a WellCare policy and procedure document agreeing to adhere to the 80/20 requirement described in the 2006 AHCA contract. 40 Case: 14-12373 Date Filed: 08/11/2016 Page: 41 of 124 In February 2007, AHCA sent Staywell and HealthEase the Worksheets for CY 2006 with instructional cover letters. The Worksheets cited the 80/20 law and explained that Staywell and HealthEase were required to spend at least 80% of their outpatient behavioral health premium money on “behavioral health services.” The Worksheets defined “behavioral health services” as “community mental health and targeted case management services only.” (emphasis added). The Worksheets were clear that AHCA was asking Staywell and HealthEase to state expenses for only CMH/TCM services. The Worksheets required the CEO or President of Staywell and HealthEase to certify that the reported expenses were true and correct. AHCA completed line 1 of the Worksheets, listing the portion of Staywell’s and HealthEase’s premium allocated to CMH/TCM services. The CY 2006 cover letters closely mirrored the CY 2005 cover letters. Like the Worksheets, the letters instructed that Staywell and HealthEase were subject to the 80/20 law and quoted a portion of the the statute as follows: To ensure unimpaired access to behavioral health care services by Medicaid beneficiaries, all contracts issued pursuant to this paragraph shall require 80 percent of the capitation paid to the managed care plan, including health maintenance organizations, to be expended for the provision of behavioral health care services. In the event the managed care plan expends less than 80 percent of the capitation paid pursuant to this paragraph for the provision of behavioral health care services, the difference shall be returned to the agency. The letters listed the specific CMH/TCM procedure codes that Staywell and HealthEase could count in reporting qualifying expenses. The letters admonished: 41 Case: 14-12373 Date Filed: 08/11/2016 Page: 42 of 124 “Report expenditures for behavioral health care services that cover targeted case management and community mental health services only.” The letters invited Staywell and HealthEase to contact AHCA if they had any questions regarding their reporting obligations. A group email exchange ensued, which included Behrens, Kale, and Clay. Behrens announced to the group that he would “take point” on completing Staywell’s and HealthEase’s 80/20 submissions. For CY 2006, Staywell and HealthEase had modified their contracts with Harmony and increased their sub- capitation rates and payments. This adjustment was intended to account for the increased premium AHCA was paying now that the CMH/TCM program was statewide. West testified, however, that he calculated the new sub-capitation rates, which had nothing to do with actual behavioral health care expenses. West set the new rates to reflect 85% of Staywell’s and HealthEase’s projected premium for CMH/TCM services. West testified that during a meeting in Behrens’s office, he related that another company had paid $5 million to settle with AHCA over the reporting method it had used. West personally hoped Behrens would “take the bait.” But Behrens explained, “[T]he system works good for us. We pay them a million dollars. That’s enough. They think the system works, and so, that’s it.” Behrens believed that, if Staywell and HealthEase refunded about one million dollars to 42 Case: 14-12373 Date Filed: 08/11/2016 Page: 43 of 124 AHCA, AHCA would likely just accept Staywell’s and HealthEase’s numbers and forgo an audit. In determining the expense figures to report for CY 2006, West worked with actuary Jian Yu, the new director of WellCare’s Medical Economics department. West explained to Yu (1) how Staywell and HealthEase had determined their expense figures in previous years and (2) that, the year before, Farha wanted to refund about one million dollars to AHCA. In West’s words, “it became ‘how do you get there.’” West told Yu of his concern that since Staywell and HealthEase had increased their sub-capitation rates and payments to Harmony, Staywell and HealthEase might not have any amount to refund to AHCA at all. Yu told West to calculate expenses the same way as he had the previous year and to get the refunds as close as he could to the CY 2005 numbers. Subsequently, West sent Yu a spreadsheet that displayed Staywell’s and HealthEase’s expense and refund figures for all prior reporting years. West’s spreadsheets also displayed three CY 2006 refund scenarios, each showing different expense figures that yielded different refund amounts. In each scenario, West used the inpatient rate from the previous year to calculate the portion of the sub-capitation payments that Staywell and HealthEase would count as qualifying CMH/TCM expenses. 43 Case: 14-12373 Date Filed: 08/11/2016 Page: 44 of 124 In the first scenario, West used the amount of the outpatient portion of Staywell’s and HealthEase’s sub-capitation payments to Harmony and reduced it by a specific sum, which West labeled a “Missing Premium.” This scenario mirrored West’s methodology for the CY 2005 Worksheets, except it did not involve double-counting both sub-capitation payments to Harmony and some of Harmony’s fee-for-services costs paid to providers. The second scenario was the same except the “Missing Premium” amount was reduced. The third scenario did not include a “Missing Premium” item at all, resulting in Staywell’s and HealthEase’s “Medical Costs” being the same hypothetical outpatient portion of the sub-capitation payments to Harmony (calculated by subtracting the inpatient portion of the sub-capitation, based on an artificial inpatient rate of $4.68 PMPM). The third scenario was similar to the methodology West used for CY 2004. West calculated the total combined refund for Staywell and HealthEase under each of these three scenarios as: (1) $1,948,246; (2) $1,354,226; and (3) $0. None of West’s scenarios attempted to calculate as qualifying expenses what Harmony had actually paid to providers of CMH/TCM services. West recommended the second scenario to Yu because it was the best option for reaching a refund between $1 million and $1.5 million. Yu disagreed, preferring not to use a “Missing Premium” calculation at all. Yu instead asked West to calculate the percentage of outpatient behavioral health care claims that 44 Case: 14-12373 Date Filed: 08/11/2016 Page: 45 of 124 used AHCA-approved CMH/TCM procedure codes and to multiply that percentage by the outpatient portion of the sub-capitation payments to Harmony. The use of the CMH/TCM codes in this way still would not generate accurate expenses because the percentage Yu asked West to generate was a percentage of total claims using the authorized codes, not a percentage of total dollars spent on authorized claims. Another serious problem with this calculation was that West did not have any current claims data, and the submission deadline was near. So with Yu’s approval, West used older claims data to generate the percentage figure Yu requested (incidentally 85%). He multiplied 85% by the outpatient portion of the sub-capitation payments to Harmony. Doing so yielded an expense percentage of 77.0% and a combined refund total of $1,108,726. West and Yu met with Behrens several times to discuss their calculations. After West and Yu finalized their calculations, Behrens asked Yu why they were not adjusting their expense figures to account for the premium difference as they had for CY 2005. Yu responded that such a method was not “actuarially sound.” In response to Yu’s comment, Behrens grinned at West, licked his thumb, and held it up, as if testing the weather. Staywell and HealthEase once again submitted to AHCA their certified Worksheets. Staywell’s Worksheet certified that Staywell spent $14,235,874 or 45 Case: 14-12373 Date Filed: 08/11/2016 Page: 46 of 124 78.3% of its premium for CMH/TCM on qualifying services in CY 2006, resulting in a $305,828 refund to AHCA. HealthEase’s Worksheet certified that HealthEase spent $14,668,012 or 75.9% of its premium for CMH/TCM on qualifying services, resulting in a $802,898 refund to AHCA. The combined total expenses for Staywell and HealthEase was $28,903,886, and the combined total refund was $1,108,726. Behrens approved Staywell’s and HealthEase’s refunds to AHCA, and the refund checks bore Farha’s signature. At trial, West testified that the expenses Staywell and HealthEase reported in their CY 2006 Worksheets were “false number[s].” Kelly, the forensic accountant, testified that Staywell’s and HealthEase’s reported expenses were “not true and accurate” and bore “[n]o logical relationship” to “moneys paid to third-party providers for the provision of outpatient behavioral healthcare services.” Based on his claims analysis, Kelly testified that Staywell and HealthEase’s combined actual qualifying expenses totaled $19,909,625, which was $8,994,261 less in expenses than the $28,903,886 in expenses they reported to AHCA. Simply put, in CY 2006 Staywell and HealthEase over-reported their expenses by almost $9 million and substantially under-paid their refunds. WellCare’s own internal records show that Staywell and HealthEase reported false, inflated expense figures to AHCA in CY 2006. West’s final spreadsheet displayed these actual qualifying expenses and corresponding refunds, 46 Case: 14-12373 Date Filed: 08/11/2016 Page: 47 of 124 along with the falsely inflated expenses and correspondingly deflated refunds Staywell and HealthEase submitted to AHCA. As both West and Kelly explained at trial, West’s spreadsheet showed that Staywell and HealthEase spent only $17,904,508 or 47.7% of their premium for CMH/TCM services on qualifying expenses, and they therefore should have refunded $12,108,104 to AHCA. Instead, Staywell and HealthEase reported $28,903,886 or 77.0% in CMH/TCM expenses and refunded only $1,108,726 in CY 2006. By CY 2006, WellCare’s use of Harmony was serving its purpose. The evidence sufficiently showed that with accurate reporting that year, Staywell and HealthEase should have refunded approximately $12 million to AHCA. But by creating Harmony and reporting what Staywell and HealthEase paid it, rather than what they paid providers of CMH/TCM services, WellCare, in CY 2006 alone, avoided refunding approximately $11 million. To avoid an audit by AHCA that might reveal this fact, Staywell and HealthEase did not even report the full sub- capitation payments that they paid Harmony. They instead manipulated the numbers to generate an arbitrary refund amount of slightly over $1 million to avoid drawing AHCA’s attention. Kelly testified that through these years, Staywell and HealthEase used inconsistent 80/20 reporting methods that started with a predetermined refund amount and worked backward to reach that result. 47 Case: 14-12373 Date Filed: 08/11/2016 Page: 48 of 124 E. Cumulative Impact Kelly testified about the cumulative impact Staywell’s and HealthEase’s use of their Harmony pass-through reporting method had on their reported expenses and refunds. He testified, based on his claims analysis, that across all reporting periods from CY 2002 to CY 2006, Staywell and HealthEase had actually paid providers only $40,805,691, which was $29,920,705 less than the $70,726,396 in total expenses they reported to AHCA. Kelly also testified that he had examined WellCare’s Form 10-K, a restated financial statement (the “restatement”) publicly filed with the Securities and Exchange Commission (“SEC”) in 2007 to correct for accounting errors in WellCare’s compliance with its refund obligations under the AHCA contracts. Kelly examined the working papers of Deloitte & Touche LLP, the outside accounting firm that audited and prepared the restatement. Based on the audited numbers in the restatement, Kelly calculated Staywell and HealthEase collectively had owed AHCA $35,134,000 more in refunds across all reporting periods than they had paid due to their false 80/20 expense reporting. Using the restatement numbers, Kelly also calculated the impact Staywell’s and HealthEase’s use of their Harmony pass-through reporting method had on WellCare’s net income before taxes for tax years 2004, 2005, and 2006. He calculated that Staywell and HealthEase’s combined net income before taxes 48 Case: 14-12373 Date Filed: 08/11/2016 Page: 49 of 124 should have been 13.9% lower in 2004, 8.8% lower in 2005, and 6.5% lower in 2006 than they had previously reported without use of their Harmony pass-through reporting method. Then, using numbers from his own claims analysis, Kelly calculated that Staywell and HealthEase’s combined net income before taxes should have been 14.7% lower in 2004, 7.4% lower in 2005, and 5.3% lower in 2006 without use of their Harmony method. Kelly testified that the two sets of figures, while not identical, nevertheless were close. He explained the utility of comparing the two sets of figures: “It’s just another measuring point to compare the results and— determine the reasonableness of my conclusion.” IV. Patient Encounter Data Between 2005 and 2007, AHCA learned of the defendants’ fraudulent 80/20 reporting through distinct but related mandatory reports. AHCA required HMOs to report data regarding encounters between patients and medical providers (patient “encounter data”). AHCA used the patient encounter data (1) to keep track of the types and frequency of medical services delivered to Medicaid patients and (2) to set future capitated rates payable to HMOs. Through 80/20 expense reporting, AHCA tracked Staywell’s and HealthEase’s annual, aggregate amounts paid to providers for CMH/TCM services. But 80/20 reporting did not reveal unit cost per 49 Case: 14-12373 Date Filed: 08/11/2016 Page: 50 of 124 service provided. In contrast, through encounter data reporting, AHCA tracked individual services provided to patients and sometimes the cost of those services. By 2005, large mismatches between Staywell’s and HealthEase’s reported 80/20 expenses and their patient encounter data reflecting unit costs for CMH/TCM services created discrepancies that AHCA investigated. AHCA requested Staywell and HealthEase to submit patient encounter data on several occasions. In earlier years, Staywell and HealthEase had priced their patient encounter data based on Harmony’s costs—that is, what Harmony paid providers for services. By 2007, they shifted to pricing their encounters based on what Staywell and HealthEase each paid to Harmony, regardless of what Harmony paid to providers. We discuss Staywell’s and HealthEase’s patient encounter data reporting because it reveals (1) the defendants’ efforts to hide from AHCA salient facts regarding their 80/20 reports and (2) the defendants’ intent to defraud with respect to the submission of those 80/20 reports. These events also bear directly on the conduct for which Clay was charged. A. Discrepancies Discovered In early 2005, AHCA discovered discrepancies between Staywell’s and HealthEase’s 80/20 expense reports and patient encounter data. Using the patient encounter data, AHCA estimated what percentage of premium for CMH/TCM 50 Case: 14-12373 Date Filed: 08/11/2016 Page: 51 of 124 Staywell and HealthEase should have spent on qualifying services. AHCA found these percentages to be far lower than the percentages Staywell and HealthEase had reported. Staywell had certified to AHCA that it spent 50.5% of its premium on CMH/TCM services in CY 2003 and 72.1% in CY 2004. HealthEase had certified that it spent 61.2% of its premium on CMH/TCM services in CY 2003 and 79.0% in CY 2004. By examining their encounter data, however, AHCA calculated that Staywell and HealthEase’s combined expenses from July 2003 through June 2004 should have totaled only 21.1% of their premium for CMH/TCM services.12 In April 2005, AHCA requested that Staywell and HealthEase provide a detailed explanation to justify the wide variance between AHCA’s 21.1% estimate and the much higher percentages Staywell and HealthEase had reported in their 80/20 expense reports. Keith Sanders, a manager in WellCare’s Medical Economics department, drafted a reply letter. The letter truthfully disclosed that, while AHCA had counted money paid to providers, Staywell and HealthEase’s 80/20 reports counted payments to Harmony: In your letter you express concern for differences between your calculated aggregate loss ratio of 21.08% and our submitted loss ratios of 72.11% and 78.99% for Staywell and HealthEase respectively. We believe the differences in loss ratio calculation are due to a difference in the view of business entity paying the costs. Our submission is 12 WellCare’s internal records for the same period reveal that Staywell and HealthEase, through Harmony, paid providers only 19.4%. 51 Case: 14-12373 Date Filed: 08/11/2016 Page: 52 of 124 based on capitated payments to Harmony Behavioral Health, Inc for the provision of covered outpatient services under the contract. Your calculation is based on capitated payments, fee for service claims, and other monthly fixed fees for the same services paid by our contracted behavioral health provider Harmony Behavioral Health, Inc to their contracted “downstream” providers. (emphasis added). On May 27, 2005, Pearl Blackburn forwarded a copy of Sanders’s draft letter to Behrens, Kale, and Clay, among others. Kale sent Behrens an email stating, “Paul, I would recommend that you or Thad [Bereday] have input in this letter. Basically, I would suggest that we again state what we did . . . without getting into much detail.” Behrens wrote back, “I agree that we need to further edit this letter.” The letters Staywell and HealthEase ultimately sent to AHCA were tight- lipped. The revised letters wholly omitted Sanders’s explanation that Staywell and HealthEase counted their sub-capitation payments to Harmony as their 80/20 expenses, without regard to how much money Harmony paid to actual providers. Staywell and HealthEase responded with a smokescreen and did not disclose the true cause of the wide variance between their reported expense percentages and AHCA’s estimate. It is unquestionable that by 2005, WellCare executives, including Behrens, Kale, and Clay, knew the wide variance was due to Staywell’s and HealthEase’s having reported their payments to Harmony rather than payments to providers. In 52 Case: 14-12373 Date Filed: 08/11/2016 Page: 53 of 124 addition to falsely reporting 80/20 expenses, by 2005 Behrens, Kale, and Clay knew that WellCare was actively misleading AHCA regarding the false reporting. B. WellCare Inflates Costs On January 2, 2007, AHCA requested Staywell and HealthEase to submit patient encounter data for behavioral health care services in Areas 1 and 6 for the period of July 1, 2005, through June 30, 2006. On January 16, 2007, Robert Butler, WellCare’s Director of Medicaid Policy Analytics and former Bureau Chief of AHCA’s Medicaid Program Analysis, convened a meeting with other WellCare employees to discuss how to price Staywell’s and HealthEase’s behavioral patient-provider encounters. Unbeknownst to the meeting’s attendees, WellCare’s Sean Hellein had begun secretly recording internal company conversations in preparation for filing a whistleblower suit. At the meeting, Butler suggested that Staywell and HealthEase price their behavioral health patient encounters to reflect what Harmony paid providers for health care services. Specifically, Butler pointed out that (1) Harmony was part of WellCare, meaning that Harmony’s overhead and profit was retained by WellCare as a whole, and thus (2) Staywell’s and HealthEase’s patient encounter data pricing should not reflect their payments to a related party (i.e. sub-capitation money paid to Harmony) but should instead reflect the cost of services (i.e. money Harmony 53 Case: 14-12373 Date Filed: 08/11/2016 Page: 54 of 124 paid to medical providers). 13 Butler asked whether Harmony provided any mental health services itself. “No,” answered one of the meeting’s attendees, “[Harmony does] utilizational review . . . it’s administrative dollars . . . [i]t’s all of our salaries.” Another added, “It’s overhead.” During the meeting, Kale and Clay entered the room and listened to Butler’s suggestion that WellCare price its patient encounter data to reflect Harmony’s payments to medical providers rather than the sub-capitation sums that Staywell and HealthEase paid Harmony. “[I]f we provide what you’re asking for,” Clay chimed in, “we’re in deep trouble.” “The whole argument for Harmony,” Clay explained to Butler, “is 85 percent, that’s our cost . . . . [T]he state is doin’ this as another end around, to find out how much money we’re makin’ in that. Which we’ve been finessing, for years.” He added, “[W]e’re gonna have huge numbers and were [sic] gonna get a massive rate cut.” Clay continued, “[P]rofit within Harmony is upwards of 50%, of that 85%. It’s huge . . . . Harmony direct expense for salaries and payroll they’d probably take it. It’s this big slug in the middle, which is, the whole reason Harmony exists, to hide this. So, are we gonna report that, or not?” 13 Later in the meeting, Butler explained why pricing encounters to reflect related-party transactions was problematic. He explained that “because it’s a cap, it’s . . . a related entity, you may be very healthy in your capitation rate.” West translated at trial: “Very healthy means . . . internally inflated costs.” In other words, because Staywell, HealthEase, and Harmony were all owned by WellCare, they had an incentive to pay Harmony much more in sub-capitation money than an unaffiliated BHO in order to internalize any money not paid out to health care providers. 54 Case: 14-12373 Date Filed: 08/11/2016 Page: 55 of 124 Clay candidly expressed his concern, which others shared, that if AHCA learned how much WellCare was profiting off of the premium for CMH/TCM services, AHCA would reduce Staywell’s and Healthease’s capitated rates. Clay continued, “Every year we’ve fed the gods. We’ve paid them a little money to keep them happy. We’ve paid them a million bucks a year, or whatever. If they’re now askin’ for us to pay it all, then let’s . . . get that conversation on the table.” Another meeting participant, Marc Ryan, shared Butler’s concern with reporting Medicaid patient encounter prices to reflect what Staywell and HealthEase paid Harmony. Ryan explained why encounters priced that way would not “sit well” with AHCA and that AHCA was expecting patient encounters to be priced at something closer to Harmony’s actual costs to providers so as to create a reliable process for setting capitated rates. But Kale disapproved of any patient encounter data pricing methodology that would reflect costs as anything less than what Staywell and HealthEase paid Harmony because WellCare had not disclosed its 80/20 reporting methodology to AHCA: While, we’ve danced around this, and we send ’em a check every year, we never, have formally been asked to justify, or we’ve never been audited for this. So we’ve never shot the [Harmony] gun ever. We’ve never had to publically say, this is how we priced it, this was our methodology, and we have [Harmony] in the middle getting 85%, and that’s where we stand. 55 Case: 14-12373 Date Filed: 08/11/2016 Page: 56 of 124 (emphasis added). After more back-and-forth, Clay added, “I don’t believe you can disconnect these [the two reporting processes] . . . . [I]f you price [the encounter data at] anything reasonable, we’re gonna show a 50% loss ratio, and we’re right back to opening the Kimono.” At trial, West explained that “[o]pening the Kimono” was to “reveal” that “WellCare should be making a huge payback” to AHCA (since Staywell’s and HealthEase’s medical costs were 40-50% as opposed to the 80% required by the 80/20 rule). Clay proposed that they calculate patient encounter unit prices by dividing the total sub-capitation paid to Harmony by the total number of encounters, which Kale supported. Doing so would allow them to account for all the sub-capitation payments to Harmony. While Butler entertained this proposal, he stressed the importance of being forthright with AHCA about it. Butler explained why patient encounter prices should reflect only actual costs in money paid to providers, not administration, overhead, and profit for Harmony: AHCA’s actuaries already built administration, overhead, and profit into the capitated rate. Butler emphasized that if they wanted to price their patient encounter prices to reflect Staywell’s and HealthEase’s sub-capitation payments, which he suggested was an “aggressive stance,” they should put a “disclaimer with it,” explicitly disclosing how they priced their patient encounters. Then, he explained, “If they don’t like the prices, they are perfectly capable of repricing them however way they want. And, we 56 Case: 14-12373 Date Filed: 08/11/2016 Page: 57 of 124 haven’t hidden anything we just told them, this is, we recognize our subcap arrangement, period.” Apparently, Butler’s recommendation of candid disclosure fell flat. As the group continued to discuss, Clay reminded the group: “The problem is we got a high margin business we are trying to protect.” Clay favored reporting their patient encounter prices to match the sub-capitation payments because doing so would put the “onus” on AHCA to negotiate the next capitated rate. Clay explained, in his view, the patient encounter data reporting process was as “much a political negotiation . . . as it [was] an analytic negotiation.” He added, “There’s more to this, than just pure analytics.” Ultimately, the group decided to price patient encounters by spreading the sub-capitation payments across all Medicaid patient encounters. The result was that Staywell and HealthEase would report prices well above Harmony’s actual costs for patient services. On January 29, 2007, in another secretly-recorded company conversation, several WellCare employees discussed the details of Staywell’s and HealthEase’s upcoming patient encounter data submissions. Clay said, “I keep wanting . . . to make this a simple conversation. It is a simple conversation. I think we’re going to have to put some numbers that are about 40 percent higher than we think they should be, because we’re making about a 40 percent profit margin. And that’s what we’re gonna submit . . . . And that’s all there is to this conversation. It’s that 57 Case: 14-12373 Date Filed: 08/11/2016 Page: 58 of 124 simple.” Clay later added, “[I]t’s just a matter of how inflated a unit cost number we’re going to be submitting.” On February 9, 2007, several WellCare employees met in Behrens’s office to discuss final matters before Staywell and HealthEase submitted their patient encounter data to AHCA. During the conversation, Bereday expressed concern about an email Butler had sent in connection with the encounter data reporting process. Bereday was concerned that Butler had carelessly conceded too much by suggesting in an email that it would be “misleading” to characterize Harmony as a provider. Sean Hellein quickly corrected Bereday: “[D]o you understand why they made that distinction? . . . . [Harmony] is not a provider.” Behrens agreed, “Uh, that’s right [Harmony] is not a provider of behavioral health services.” Behrens explained, “You can’t refer to them as a provider because technically under the, I’ll say, and maybe it’s not the law, but . . . of what, the state would consider to be a provider, is like somebody that has a license to provide medical services.” But, he added, “[Harmony] is not licensed to provide medical services.” Behrens and Hellein agreed that AHCA was concerned with actual health care services. Harmony did not provide such services and therefore was not a provider because, as Behrens put it, “[Harmony] doesn’t do the laying on of hands.” Bereday pressed, “Okay. But it’s a provider to us.” Behrens agreed in a 58 Case: 14-12373 Date Filed: 08/11/2016 Page: 59 of 124 qualified sense: “A provider of services. Just as the electric company is a provider to us.” Later that day, WellCare submitted its patient encounter data for Areas 1 and 6. Its cover letter accompanying its encounter data stated vaguely, “Mental health encounters have been priced based upon the plans’ arrangements for behavioral health services, including those paid on a capitated basis.” The cover letter still did not mention that Staywell and HealthEase priced their Medicaid patient encounters to reflect payments to Harmony rather than to providers of services, even though Butler had originally suggested that Staywell and HealthEase be forthright about this fact in their encounter data submissions. At trial, West explained that Behrens did not want that detail slipped to AHCA. Behrens even suggested that they hold a meeting after submitting their patient encounter data “to make sure that young Robert is on message.” West testified that he understood Behrens to mean that Robert Butler needed to “understand[] that the encounters [had] been priced up to [Harmony] but he’s not to reveal to the agency the relationship between HealthEase and Staywell and [Harmony] and the providers.” C. AHCA Requests Backup On April 17, 2007, after Staywell and HealthEase submitted their CY 2006 80/20 expense reports, Hazel Greenberg of AHCA emailed Butler. “Thank you for the filing of the Behavioral 80/20 refund reports and checks,” she said. “The 59 Case: 14-12373 Date Filed: 08/11/2016 Page: 60 of 124 Agency is requesting that HealthEase and [Staywell] submit the encounter data, with codes and reimbursement amounts for each code, for documentation for the 2006 Community Mental Health and Targeted Case Management Expenses.” Butler promptly alerted Behrens and Kale, among others. When West learned that AHCA “want[ed the] backups” to the 80/20 submission, “[d]own to every . . . [p]rocedure code,” he told his colleagues, “[T]he encounters aren’t gonna get you there.” “[It] goes back to where Paul [Behrens] was,” he added. “[I]f we cut ’em a check this big, they won’t do anything . . . . [W]hen they do something, that’s when you gotta pay the piper.” A colleague responded, “We should have sent them 2 million.” This was the first time AHCA had requested patient encounter data from Staywell and HealthEase as backup for their 80/20 expense reports. West was concerned because AHCA was asking for expenses paid per claim and per Medicaid patient encounter, but Staywell and HealthEase had not reported their 80/20 expenses based on actual costs in money paid to Medicaid providers. On April 19, 2007, Behrens convened a meeting with Yu and West to discuss AHCA’s request for supporting data. Behrens wanted to include as many patient encounters and procedure codes as possible, but West favored including only encounters with procedure codes expressly authorized in the cover letters accompanying the CY 2006 Worksheets. West also suggested to Behrens that 60 Case: 14-12373 Date Filed: 08/11/2016 Page: 61 of 124 Staywell and HealthEase tell AHCA that they counted the sub-capitation payments to Harmony as their expenses in their 80/20 reports. West was “shocked” at how dismissive Behrens was of that idea. Ultimately, Behrens’s team settled on including as many patient encounters as possible and including procedure codes that the Worksheet cover letters had not authorized. Staywell and HealthEase submitted their encounter data unpriced. This way, if AHCA disapproved of any procedure codes, there would not be identifiable amounts per procedure code for which AHCA might demand a refund. As a result, Staywell and HealthEase’s patient encounter data reporting methodology was inconsistent with their CY 2006 80/20 expense reporting methodology. For CY 2006, Staywell and HealthEase purportedly did not count procedure codes beyond those authorized by the Worksheet cover letters, but now Behrens ordered that those same previously-omitted codes be included. West testified that he disagreed with Staywell and HealthEase’s approach and that he expressed his concern to Behrens, whom West described as the ultimate decision- maker for the encounter data. In response, Behrens assured West that AHCA was “just going to ask for encounters and [AHCA was] going to put it on a shelf.” 61 Case: 14-12373 Date Filed: 08/11/2016 Page: 62 of 124 Behrens also told West that he “hope[d] the law would come off the books”—that is, the “80/20 law.”14 Behrens had West draft a letter in reply to AHCA regarding its patient encounter data request, the content of which Behrens and Yu dictated. West’s letter explained: We have stated in our Financial Worksheet for the Calculation of Behavioral Health Care Ratio for calendar year 2006 that Community Mental Health and Targeted Case Management Expenses are contracted on a comprehensive basis. Since the Healthease and StayWell amount paid is not determined by the encounters submitted we have not used a pricing method that would force agreement to our comprehensive payment. It should be noted that not all encounters have been received for calendar year 2006 and some providers have not forwarded all encounters due which is still in resolution at this date. West’s letter did not disclose that Staywell and HealthEase had included unauthorized procedure codes in their patient encounter data. More significantly, the letter failed to disclose Staywell and HealthEase’s use of their Harmony pass- through reporting method in their 80/20 reports. D. AHCA Requests Corrections On June 22, 2007, AHCA’s David Starn emailed Kale and explained that “the data submitted for Healthease and Staywell for the 2006 80/20 Annual 14 WellCare representatives, including Farha, lobbied lawmakers and other government officials to repeal the 80/20 law. Those efforts from 2002-2007 proved fruitless. Years later, effective June 30, 2015, Florida repealed the 80/20 provision from its Medicaid statute. See 2015 Fla. Laws 84, 87 (codified as amended at Fla. Stat. § 409.912 (2015)). 62 Case: 14-12373 Date Filed: 08/11/2016 Page: 63 of 124 Behavioral Health Expenditure report contained many procedure codes and revenue center codes that are not in our list of valid values for behavioral health reporting.” Starn added, “Most importantly, there is no Amount Paid for any of the encounters reported.” (emphasis added). Starn requested that Staywell and HealthEase resubmit their encounter data with correct information. On June 25, 2007, West alerted Behrens and Yu to Starn’s request. Later that day, in another secretly-recorded conversation, Kale, West, and several others discussed how Staywell and HealthEase should respond to Starn’s message. West explained the problem to his colleagues: “Paul [Behrens] wanted me to count everything in the encounters. But, our payback was based on not counting everything. So we had a little over a million dollars to pay back. But they thought that was, that would satisfy the AHCA gods, and it didn’t.” Kale commented, “[W]e put stuff in there [the encounter data] that we didn’t even, uh, support with our payback.” Kale expressed his concern that once AHCA was able to see what Harmony was actually paying providers and actually spending on Medicaid patient encounters, AHCA would likely reduce the premium money flowing to Staywell and HealthEase and accordingly to Harmony. Kale further commented, “Once it goes away, it’s sure gonna hurt [Harmony’s] income statement.” Kale later added, “I think the party’s over.” West explained that he could not send patient encounter data back to AHCA without first walking it past 63 Case: 14-12373 Date Filed: 08/11/2016 Page: 64 of 124 Behrens because Behrens was “the ultimate decision maker” and had “been a decision maker from the beginning.” The group also discussed a range of related issues involving the 80/20 expense reports throughout the years. Kale mentioned that he had been involved with Staywell’s and HealthEase’s 80/20 reporting for five years, and that every year “[t]he plan is give ’em [AHCA] a something . . . . Throw them a bone.” But as to whether Staywell and HealthEase had ever been up front with AHCA about their reporting methodology, Kale admitted, “[U]ltimately we haven’t formally said, oh, well we have [Harmony].” After the meeting, Kale emailed a Harmony employee and explained, “[West] is going to start re-pricing the encounters.” Kale added, “I think this ultimately will lead to Paul Behrens, Thad [Bereday] and possibly Todd [Farha] weighing in on the strategy to take with AHCA since the dollar difference is $7-10M.” West began re-pricing Staywell’s and HealthEase’s patient encounter data. This time, West used only authorized procedure codes. West “priced up” all of the Medicaid patient encounters to at least match the 85% premium money Staywell and HealthEase had paid Harmony as expenses submitted in their 80/20 reports for CY 2006. As with the earlier submission for Areas 1 and 6, this method allowed West to evenly spread Staywell’s and HealthEase’s reported 80/20 expenses across all of their qualifying patient encounters. West testified that, as Behrens described 64 Case: 14-12373 Date Filed: 08/11/2016 Page: 65 of 124 it, West “[s]pread it like peanut butter, spread it across everything.” As a result, Staywell’s and HealthEase’s patient encounter data was again false and did not reflect unit costs of Medicaid patient encounters—that is, money paid to Medicaid providers—which AHCA was obviously requesting. WellCare resubmitted Staywell’s and HealthEase’s patient encounter data as back-up for their CY 2006 80/20 expenses. In a letter accompanying the submission, WellCare failed to disclose that Staywell and HealthEase were reporting Medicaid patient encounters based on the payments to Harmony rather than on the money Staywell and HealthEase (through Harmony) paid providers.15 E. Raid on WellCare and Clay’s False Statements On October 24, 2007, over 200 federal investigators raided WellCare’s corporate headquarters in Tampa and executed a search warrant of the premises. During the raid, Clay agreed to be interviewed by two federal investigators, FBI Agent Vic Milanes and Agent Blair Johnston of the U.S. Department of Health and Human Services. The agents interviewed Clay in his office for approximately an hour and a half, discussing issues related to Staywell’s and HealthEase’s 80/20 reports. Agent Milanes asked Clay questions. Agent Johnston later memorialized 15 The government’s brief marshals Rule 404(b) evidence presented at trial showing Behrens and Clay’s participation in a false expense reporting scheme under a separate Florida statute, governing the Florida Healthy Kids program. Because the evidence is more than sufficient to sustain their convictions for their roles in producing Staywell’s and HealthEase’s fraudulent expense reports, we need not expand this opinion to set forth this Rule 404(b) evidence. On appeal, no one argues that this Rule 404(b) evidence was wrongfully admitted. 65 Case: 14-12373 Date Filed: 08/11/2016 Page: 66 of 124 the details of the interview in a report. While the notes Agent Johnston took of Clay’s responses were not verbatim, Agent Johnston testified that he attempted to use Clay’s own words. Agent Milanes asked Clay if Staywell and HealthEase had over-reported their outpatient behavioral health costs to AHCA over the years in order to avoid paying money back to AHCA. Clay responded that, to his knowledge, they had not. Agent Milanes also asked Clay whether Staywell and HealthEase had purposefully inflated the costs of their behavioral health encounter submissions to AHCA. Clay responded that, to his knowledge, they had not. Agent Milanes then asked Clay whether he had ever attended a meeting where it was discussed or suggested that Staywell and HealthEase should inflate the unit costs of their encounter claims over the actual costs in their submissions to AHCA. Clay answered that there had been no intentional inflation of costs discussed at meetings concerning AHCA’s encounter or claims information requests. At trial, Agent Johnston testified that he did not recall Clay asking for clarification of any questions Agent Milanes asked him. After the raid, Kale told West, “[Y]ou have nothing to worry about . . . . I may have something to worry about, but you have nothing to worry about.” Kale also reached out to Pearl Blackburn. Kale told Blackburn that “he had made up 66 Case: 14-12373 Date Filed: 08/11/2016 Page: 67 of 124 numbers.” When Blackburn asked why Kale would do that, Kale said that “he thought he could get away with it” and “that it was a game.” With this factual background, we now consider the issues on appeal. V. SUFFICIENCY OF THE EVIDENCE As to the CY 2006 expense reports, defendants Farha, Behrens, and Kale challenge the sufficiency of the evidence as to their § 1347 convictions for health care fraud and defendant Behrens does also as to his § 1035 convictions for making false representations to AHCA. Clay separately challenges his § 1001 convictions for making false statements to federal agents. All defendants contend that the district court erred in denying their motions for judgment of acquittal. A. Standard of Review We review de novo a district court’s denial of a Rule 29 motion for judgment of acquittal, “viewing the evidence in the light most favorable to the government and drawing all reasonable inferences in favor of the jury’s verdict.” United States v. Martin, 803 F.3d 581, 587 (11th Cir. 2015). “‘The test for sufficiency of the evidence is identical, regardless of whether the evidence is direct or circumstantial,’ but if the government relied on circumstantial evidence, ‘reasonable inferences, not mere speculation, must support the conviction.’” Id. (citation and alterations omitted). 67 Case: 14-12373 Date Filed: 08/11/2016 Page: 68 of 124 “It is not enough for a defendant to put forth a reasonable hypothesis of innocence, because the issue is not whether a jury reasonably could have acquitted but whether it reasonably could have found guilt beyond a reasonable doubt.” United States v. Thompson, 473 F.3d 1137, 1142 (11th Cir. 2006). “We will not overturn a jury’s verdict if there is ‘any reasonable construction of the evidence that would have allowed the jury to find the defendant guilty beyond a reasonable doubt.’” Martin, 803 F.3d at 587 (alterations omitted). The jury has exclusive province over the credibility of witnesses, and we may not revisit the question. United States v. Hernandez, 743 F.3d 812, 814 (11th Cir. 2014). B. Health Care Fraud Under §§ 1347 and 1035 Farha, Behrens, and Kale were convicted of health care fraud committed in CY 2006, in violation of 18 U.S.C. §§ 1347 and 2 (Counts 8 and 9). Section 1347 makes it a crime for an individual “knowingly and willfully” to execute, or attempt to execute, a scheme or artifice “(1) to defraud any health care benefit program” or “(2) 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” if done “in connection with the delivery of or payment for health care benefits, items, or services.” 18 U.S.C. § 1347(a). Section 1347(a) proscribes: (1) fraud on a health care benefit program, here the Florida Medicaid program, see 18 U.S.C. § 1347(a)(1); and (2) obtaining a 68 Case: 14-12373 Date Filed: 08/11/2016 Page: 69 of 124 program’s money “by means of false or fraudulent . . . representations,” see id. § 1347(a)(2); accord United States v. Dennis, 237 F.3d 1295, 1303 (11th Cir. 2001) (noting that an offense under the similarly-structured and similarly-worded bank fraud statute, 18 U.S.C. § 1344, “is established under two alternative methods”) (citing United States v. Goldsmith, 109 F.3d 714, 715 (11th Cir. 1997)). The indictment charged Farha, Behrens, and Kale with both types of health care fraud covered by § 1347. The core fraudulent conduct was generally similar for both. Specifically, the defendants participated in a scheme to defraud AHCA by submitting, or aiding and abetting the submission of, false expense amounts in the CY 2006 Worksheets in order to reduce their AHCA refunds by millions of dollars. The government thus had to prove that (1) the CMH/TCM expenses reported in the CY 2006 Worksheets submitted to AHCA were, in fact, false; and (2) the defendants knew those representations were, in fact, false. See United States v. Vernon, 723 F.3d 1234, 1273 (11th Cir. 2013) (citing United States v. Medina, 485 F.3d 1291, 1297 (11th Cir. 2007)). Behrens was also convicted of making false and fraudulent representations in matters involving a health care benefit program, in violation of 18 U.S.C. §§ 1035 and 2 (Counts 4 and 5). Section 1035 makes it a crime for an individual, “in any matter involving a health care benefit program,” to “knowingly and willfully” (1) falsify, conceal, or cover up by any trick, scheme, or device a 69 Case: 14-12373 Date Filed: 08/11/2016 Page: 70 of 124 material fact or to (2) make any materially false, fictitious, or fraudulent statements or representations, or make or use any materially false writing or document knowing the same to contain any materially false, fictitious, or fraudulent statement or entry, in connection with the delivery of or payment for health care benefits, items, or services.16 U.S.C. § 1035(a). The indictment charged Behrens under § 1035(a)(2) for making, or aiding and abetting the making of, materially false, fictitious, and fraudulent representations. The core fraudulent conduct was similar to that charged under § 1347. The government had to prove that the CMH/TCM expenses reported in the CY 2006 Worksheets were, in fact, false and Behrens knew that they were, in fact, false. Furthermore, Farha, Behrens, and Kale were charged under an aiding and abetting theory in their § 1347 health care fraud counts and so too was Behrens in his § 1035 false representation counts. Regardless of who principally executed the fraud in CY 2006 or signed the CY 2006 expense reports, the defendants could be convicted if they aided, abetted, counseled, induced, or procured the commission of the false representations, or if they willfully caused the false representations to be committed. United States v. Sosa, 777 F.3d 1279, 1292 (11th Cir. 2015) (citing 16 The term “health care benefit program” has the same meaning in § 1035 as it does for purposes of § 1347. 18 U.S.C. §§ 23(b), 1035(b). The parties do not contest that the Florida Medicaid program administered by AHCA meets the definition of “health care benefit program.” 70 Case: 14-12373 Date Filed: 08/11/2016 Page: 71 of 124 18 U.S.C. § 2). “Under 18 U.S.C. § 2, aiding and abetting is not a separate federal crime, ‘but rather an alternative charge that permits one to be found guilty as a principal for aiding or procuring someone else to commit the offense.’” Id. C. CY 2006 Reported Expenses Were False On appeal, Farha, Behrens, and Kale primarily contend that (1) the expense amounts for CMH/TCM services to Medicaid patients, as reported in the CY 2006 Worksheets, were true, not false, and, in any event, (2) they did not know that those reported expense amounts were false. Our extensive review of the evidence above allows for brevity in this analysis. Abundant evidence established that Staywell and HealthEase reported false and fraudulent CY 2006 expenses. Staywell and HealthEase never reported the amounts paid to providers of CMH/TCM services to Medicaid patients or even the accurate sums paid to Harmony. Both West, WellCare’s own employee, and Kelly, the forensic accountant, testified that the reported CMH/TCM expense amounts were false and explained why. In the raid, the government obtained WellCare’s own internal records that showed exactly what total expense amounts were paid to providers, and those amounts were millions below what Staywell and HealthEase reported to AHCA. Defendants claim their CMH/TCM expense reports were truthful because the 80/20 rule did not require them to report money paid to health care providers of 71 Case: 14-12373 Date Filed: 08/11/2016 Page: 72 of 124 CMH/TCM services, but allowed them to report what was paid to Harmony. Defendants’ arguments fail for multiple reasons. First, AHCA asked and required Staywell and HealthEase to report what they paid providers of CMH/TCM services—not companies (like Harmony) that rendered administrative services. The defendants rely on the language of Florida’s 80/20 law, but Staywell’s and HealthEase’s reporting obligations were governed not only by that law but also by (1) their 2006 contracts with AHCA, (2) the instructions included on the 80/20 Worksheets, and (3) the specific procedure codes and instructions in AHCA’s cover letters accompanying the 80/20 Worksheets. Read together, nothing was ambiguous about what Staywell and HealthEase were required to report on line 2 of the CY 2006 Worksheets. The 80/20 law was clear. To ensure access to care for Medicaid patients, the 80/20 law mandated that all of AHCA’s contracts “shall require” that 80% of the premium paid to a health plan must be expended for behavioral health care services: To ensure unimpaired access to behavioral health care services by Medicaid recipients, all contracts issued pursuant to this paragraph shall require 80 percent of the capitation paid to the managed care plan, including health maintenance organizations, to be expended for the provision of behavioral health care services. In the event the managed care plan expends less than 80 percent of the capitation paid pursuant to this paragraph for the provision of behavioral health care services, the difference shall be returned to the agency. 72 Case: 14-12373 Date Filed: 08/11/2016 Page: 73 of 124 Fla. Stat. § 409.912(4)(b) (2006) (emphasis added). The statute made explicit that if Staywell and HealthEase expended less than 80% of their premium “for the provision of behavioral health care services,” then “the difference shall be returned to [AHCA].” Id. Likewise, the 2006 AHCA contract was clear. The contract included an entire section, titled “Community Behavioral Health Services Annual 80/20 Expenditure Report,” explaining Staywell’s and HealthEase’s 80/20 reporting obligations.17 The section informed Staywell and HealthEase that 80% of their premium shall be expended for behavioral health care services, as follows: 1. By April 1 of each year, Health Plans shall provide a breakdown of expenditures related to the provision of community behavioral health services, using the spreadsheet template provided by the Agency (see Section XII, Reporting Requirements). In accordance with Section 409.912, F.S., eighty percent (80%) of the Capitation Rate paid to the Health Plan by the Agency shall be expended for the provision of community behavioral health services. In the event the Health Plan expends less than eighty percent (80%) of the Capitation Rate, the Health Plan shall return the difference to the Agency no later than May 1 of each year. a. For reporting purposes in accordance with this Section, ‘community behavioral health services’ are defined as those services that the Health Plan is required to provide as listed in the Community Mental Health Services Coverage and Limitations Handbook and the Mental Health Targeted Case Management Coverage and Limitations handbook. 17 The defendants argue that the 2006 AHCA contract, rather than any of its prior versions, is the relevant contract for purposes of determining what Staywell’s and Healthease’s 80/20 reporting obligations were for CY 2006. The government does not disagree. We therefore consider the 2006 contract for purposes of our analysis. 73 Case: 14-12373 Date Filed: 08/11/2016 Page: 74 of 124 Most importantly, the section expressly and precisely described qualifying expenses under the 80/20 rule. The section explained that “expended” meant (1) the money paid to “community behavioral health services providers solely for the provision” of CMH/TCM services and (2) did not include “administrative expenses or overhead of the plan,” stating: b. For reporting purposes in accordance with this Section ‘expended’ means the total amount, in dollars, paid directly or indirectly to community behavioral health services providers solely for the provision of community behavioral health services, not including administrative expenses or overhead of the plan. If the report indicates that a portion of the capitation payment is to be returned to the Agency, the Health Plan shall submit a check for that amount with the Behavioral Health Services Annual 80/20 Expenditure Report that the Health Plan provides to the Agency. (emphasis added). Under the transparent, unambiguous language of the statute and the 2006 contract, Staywell and HealthEase could count money paid to providers, but could not count administrative expenses or overhead. The Worksheets and cover letters reinforced the contract’s reporting requirements and also cited the 80/20 law. They instructed that at least 80% of the premium had to be expended on behavioral health care services, defined as community mental health services and targeted case management services. The instructions in the letters even listed the precise “procedure codes” for those health 74 Case: 14-12373 Date Filed: 08/11/2016 Page: 75 of 124 care services. Each procedure code was tied to a medical service and was not linked to any administrative or overhead expenses. Together, the 80/20 law, the 2006 contract, the Worksheets, and the cover letters posed an unmistakable question to Staywell and HealthEase: What amount of money did you pay to providers for their CMH/TCM services to Medicaid patients? They answered that question falsely. A truthful answer would have caused Staywell and HealthEase to pay large refunds to AHCA. Further undermining the defendants’ argument, the amounts Staywell and HealthEase reported were not based on CMH/TCM expenses at all, whether paid to Harmony or paid to providers. The amounts on line 2 were entirely fabricated and false figures. Staywell and HealthEase did not report on line 2 the 85% sub- capitation payments to Harmony, as then no refund would be due to AHCA. To avoid an audit and AHCA’s discovery that providers were receiving only 45% of the premium for CMH/TCM services, Farha directed his employees to generate a refund to AHCA of approximately $1 million, and his subordinates then used fabricated and false numbers to create the refund amount that Farha wanted. Year after year, fictitious inpatient and outpatient rates, double counting, premium- difference machinations, and other arbitrary calculations were used to create a predetermined refund figure. In CY 2006, that amount was $1.1 million. The defendants modified line 2 based on the refund amount that Fahra wanted to “pay 75 Case: 14-12373 Date Filed: 08/11/2016 Page: 76 of 124 the Gods” to prevent an audit. Staywell’s and HealthEase’s reported figures were not based on an analysis of accurate claims data or on a misinterpretation of qualifying expenses. Rather, Staywell and HealthEase reported expenses based upon backwards, results-oriented calculations and never reported what they paid providers of CMH/TCM services to Medicaid patients. D. Whiteside Decision The defendants rely heavily on United States v. Whiteside, 285 F.3d 1345 (11th Cir. 2002). They argue in effect that, based on Whiteside, regulated industries and their executives should be protected from the improper criminalization of routine contractual and regulatory disagreements. Whiteside, however, is materially different and, if anything, undermines the defendants’ arguments. Whiteside dealt with an ambiguous regulation for categorization of debt under 42 C.F.R. § 413.153(b)(1). Id. at 1352. The Whiteside defendants were convicted of making false statements regarding loan interest in cost reports submitted to Medicare for reimbursement. Id. at 1345-46. A regulation prescribed the amount of interest a medical provider could attribute to the provider’s own capital-related costs, which were reimbursed more favorably. Id. at 1346. But the regulation did not clarify whether capital-related costs were those for which the loan money was originally used or those for which the money was presently used at the time of filing. Id. at 1351-53. The Whiteside 76 Case: 14-12373 Date Filed: 08/11/2016 Page: 77 of 124 defendants’ cost reports classified certain loan-related interest expenses as 100% capital related. Id. at 1351. The government contended the defendants’ reporting methodology violated Medicare regulations, and the defendants were convicted of conspiracy to defraud the government, in violation of 18 U.S.C. §§ 371 and 2, and making false statements in applications for Medicare benefits, in violation of 18 U.S.C. §§ 1001 and 2. Id. at 1350. This Court reversed, finding that “competing interpretations of the applicable law” governing the cost reports were “far too reasonable to justify” the defendants’ convictions. Id. at 1353. This is because “no Medicare regulation, administrative ruling, or judicial decision exist[ed] that clearly require[ed] interest expense to be reported in accordance with the original use of the loan” as opposed to the use of the loan at the time of filing. Id. at 1352. Because the Whiteside defendants submitted information based upon a reasonable interpretation of the regulations, this Court decided that the “government failed to meet its burden of proving the actus reus of the offense—actual falsity as a matter of law.” Id. at 1353. We stated that “[i]n a case where the truth or falsity of a statement centers on an interpretative question of law, the government bears the burden of proving beyond a reasonable doubt that the defendant’s statement is not true under a reasonable interpretation of law.” Id. at 1351. Additionally, there was evidence in 77 Case: 14-12373 Date Filed: 08/11/2016 Page: 78 of 124 Whiteside that the defendants genuinely believed their interpretation was correct. Id. at 1348 (noting that “[t]hey firmly believed that the interest was 100% capital- related”). In stark contrast to Whiteside, Staywell’s and HealthEase’s reporting obligations were not governed simply by the Florida 80/20 law itself. Rather, through the years, AHCA clarified and plainly set forth Staywell’s and HealthEase’s reporting obligations in their AHCA contracts, the Worksheets, and the cover letters and instructions attached to the Worksheets. In CY 2006, AHCA executed new contracts with Staywell and HealthEase, which directly instructed: “For reporting purposes . . . ‘expended’ means the total amount, in dollars, paid directly or indirectly to community behavioral health services providers solely for the provision of community behavioral health services, not including administrative expenses or overhead of the plan.” The Worksheets came with cover letters that listed the designated procedure codes for the expenses that could be included in the reports. None of these procedure codes were for the administrative services and overhead of Harmony. The defendants argue that a reasonable interpretation of Staywell’s and HealthEase’s reporting obligations was that they could report what they paid to Harmony (even though Harmony provided only administrative services for Staywell and HealthEase) rather than the roughly 45% amount Harmony paid to 78 Case: 14-12373 Date Filed: 08/11/2016 Page: 79 of 124 providers. But Harmony itself provided no CMH/TCM services to any Medicaid patients. The defendants’ interpretation ignores the plain meaning of the AHCA contracts, the Worksheets, the cover letters, and the 80/20 law itself: no less than 80% of the premium for CMH/TCM services was to be spent on the treatment of Medicaid patients. Indeed, the defendants’ interpretation would strip the “80/20” requirement in the law and the AHCA contracts of any real meaning. Given the clarity of the instructions in the 2006 contract, the Worksheets, and the cover letters containing procedure codes, we conclude that is not a reasonable legal interpretation of Staywell’s and HealthEase’s reporting obligations for CMH/TCM expenses. At any rate, a plethora of evidence established that the defendants never believed that Staywell and HealthEase could report CMH/TCM expenses this way. The defendants fully knew that what Staywell and HealthEase were reporting was not what AHCA requested. We need not further analyze the defendants’ post-hoc interpretation because, as discussed below, the evidence in the light most favorable to the jury’s verdict shows that the defendants did not believe it, knew what was required, and knew their answers were false. E. Knowledge of Falsity The evidence overwhelmingly showed the defendants well understood their CMH/TCM expense reporting obligations and knew that the CMH/TCM expense 79 Case: 14-12373 Date Filed: 08/11/2016 Page: 80 of 124 amounts reported in the 80/20 Worksheets were false. From beginning to end, the defendants’ knowledge of that falsity remained constant. We discuss the evidence first as to Behrens and Kale and then as to Farha. From the outset, Kale knew Florida’s new 80/20 law would affect WellCare’s profits. He was one of the first to warn his colleagues about it, estimating that, under the new rule, Staywell and HealthEase might collectively be required to refund almost $6.5 million in Medicaid payments. The specter of a multi-million dollar annual refund spurred Farha, Kale, and others to create a fraudulent scheme to avoid that refund. Kale knew the game plan. He personally circulated a company slide presentation containing the “Fund Allocation Model,” which showed that Staywell and HealthEase would each pass 85% of their premium along to Harmony, but Harmony would pay only 45% of the premium to providers. Kale knew WellCare had created Harmony to serve as a “conceptual pass through,” enabling Staywell and HealthEase to report CMH/TCM expenses of at least 80% and avoid a refund. Kale also knew that Harmony would no longer be necessary if Florida repealed the 80/20 law. But Florida did not, and that meant Staywell and HealthEase were required to comply with the law by annually reporting how much of the premium for CMH/TCM services was actually paid to health care providers treating Medicaid patients. That compliance task fell to Behrens. As head of Finance at WellCare, 80 Case: 14-12373 Date Filed: 08/11/2016 Page: 81 of 124 Behrens was the “owner” of the 80/20 reporting project. For the CY 2006 reporting cycle, Behrens again announced that he would “take point.” West and others on the Medical Economics team regularly met in Behrens’s office to confer, and the team could not report expenses or issue refunds to AHCA without Behrens’s approval. The Medical Economics team worked for Behrens, not the other way around. And while not formally part of the Finance Department, Kale assisted and advised the reporting project year after year. As Kale candidly remarked to some colleagues in 2007, every year “[t]he plan [was] give ‘em [AHCA] a something . . . . Throw them a bone.” So that is what they did. The defendants’ frank comments, as revealed by company emails and secretly-recorded conversations, show that they knew creating and using Harmony—to still pay medical providers only 45% and retain the rest for overhead and profits— contravened Staywell’s and HealthEase’s compliance obgligations. As Kale remarked on the eve of Harmony’s creation: “[S]etting up the corporation is easy; it is the questions that follow . . . that will determine if we create a viable organization if we were to be audited by AHCA.” Avoiding an AHCA audit became the defendants’ perennial mission. To achieve that, Farha and his team set a one-million-dollar refund target— theoretically just enough to satisfy AHCA and avoid suspicion. As Behrens 81 Case: 14-12373 Date Filed: 08/11/2016 Page: 82 of 124 explained to West in 2007: “[T]he system works good for us. We pay them a million dollars. That’s enough. They think the system works, and so, that’s it.” As this 2007 exchange reveals, the defendants sought to avoid any interaction with AHCA that might disclose Staywell and HealthEase’s fraudulent reporting methodology. Behrens repeatedly rejected any suggestion that WellCare contact AHCA about 80/20 or encounter data reporting. As early as 2005, Behrens knew precisely why there was a large variance between AHCA’s estimate of Staywell’s and HealthEase’s CMH/TCM expenses and their reported expenses. But rather than respond to AHCA’s inquiries with a forthright disclosure of their reporting method as Sanders suggested, Behrens and Kale vetoed Sanders’s letter and instead perpetuated the fraudulent scheme. Behrens and Kale knew they were misleading AHCA with Staywell’s and HealthEase’s 80/20 and encounter data reporting. As Kale admitted: “[W]e’ve never shot the [Harmony] gun ever. We’ve never had to publically say, this is how we priced it, this was our methodology, and we have [Harmony] in the middle getting 85%.” The defendants made sure to keep it that way as long as they could. And the reason was obvious. As Behrens explained to a colleague in 2007, “[Harmony] is not a provider of behavioral health services.” That is why Behrens and his colleagues hoped the 80/20 law “would come off the books.” 82 Case: 14-12373 Date Filed: 08/11/2016 Page: 83 of 124 The defendants knew if AHCA realized that their CMH/TCM expenses were not nearly as high as they reported, more refunds would be owed and AHCA would later reduce the premiums too. And so year after year, including in CY 2006, although they knew Harmony was not a provider of health care services to Medicaid patients, they continued to report CMH/TCM expenses far in excess of their actual incurred expenses for CMH/TCM services. As the evidence shows, the defendants knew Staywell and HealthEase did not even report their full sub- capitation payments to Harmony, opting instead for a lesser amount through unsound, results-oriented accounting techniques to settle on an inconspicuous refund. The evidence amply showed that the representations as to CMH/TCM expenses in the CY 2006 expense reports submitted to AHCA were, in fact, false, and the that defendants knew they were, in fact, false. See Vernon, 723 F.3d at 1273. The evidence was more than sufficient to sustain Behrens’s and Kale’s convictions for Medicaid health care fraud, in violation of 18 U.S.C. § 1347 (Counts 8 and 9), and Behrens’s separate convictions for false representations relating to health care matters, in violation of 18 U.S.C. § 1035 (Counts 4 and 5). 18 18 Behrens also argues Counts 4 and 5 of the indictment failed to allege the essential facts of the crime. We reject Behrens’s additional argument that the district court erred in denying his motion to dismiss Counts 4 and 5 in the indictment. 83 Case: 14-12373 Date Filed: 08/11/2016 Page: 84 of 124 F. Farha’s Role Farha further challenges his § 1347 convictions, contending that (1) he played no role whatsoever in preparing, reviewing, or approving the CY 2006 expense reports, and (2) even if he did play a role, the government failed to prove the criminal intent required to impose criminal liability for health care fraud under § 1347. As Farha notes, under § 1347, the government must show that the defendant “knowingly and willfully” executed or attempted to execute the fraud. 18 U.S.C. § 1347(a). A defendant acts willfully when he acts with “knowledge that his conduct was unlawful” and acts knowingly if he acts with “knowledge of the facts that constitute the offense.” United States v. Dominquez, 661 F.3d 1051, 1068 (11th Cir. 2011). In this case, the district court instructed the jury that it must also find that the defendants acted with “intent to defraud,” defined as “specific intent to deceive or cheat someone and to deprive someone of money or property.” See United States v. Klopf, 423 F.3d 1228, 1240 (11th Cir. 2005). And as we have already explained, with health care fraud charges premised on false and fraudulent representations, “the defendant must be shown to have known that the claims submitted were, in fact, false.” Vernon, 723 F.3d at 1273. In distilling his various arguments, we observe that Farha primarily invites us to close our eyes to all evidence of his conduct outside the narrow window of 84 Case: 14-12373 Date Filed: 08/11/2016 Page: 85 of 124 time during which Behrens’s team prepared the CY 2006 expense reports. But the CY 2006 reporting cycle did not occur in a vacuum. In the 80/20 reports for CY 2006, Staywell and HealthEase continued the scheme that Farha set up in prior years, using Harmony to fraudulently report inflated and false CMH/TCM expenses. The evidence showed that Farha, as CEO, President, and a WellCare director, designed and implemented the scheme specifically to defraud AHCA and ordered his subordinates under his authority to perpetuate the scheme year after year, including CY 2006. Farha was fully aware of how the 80/20 rule affected WellCare’s bottom line, thanks in part to the profitability and refund studies actuary Todd Whitney produced. Farha hatched a plan to avoid the 80/20 rule’s effects. That plan started with the creation of Harmony, WellCare’s new wholly-owned subsidiary. Farha kept regular contact with his team during the summer and fall of 2003. He stayed informed of the Harmony project’s progress and sent emails to subordinates rebuking them for moving too slowly. The initial plan, as Farha instructed, was that Harmony would “be capped at 80% of premium.” Frustrated with his team’s slow progress, Farha ordered Kale to ensure that Harmony was up and running as soon as possible. Farha asked, “Why would we delay and increase the amount of our potential giveback?” 85 Case: 14-12373 Date Filed: 08/11/2016 Page: 86 of 124 Once Harmony was incorporated, Farha became Harmony’s President, CEO, and director-chairman. Once Harmony was up and running and after the first round of 80/20 expense reporting, Farha instructed Kale to have the subsidiary company’s name changed from its original name of “WellCare Behavioral Health, Inc.” to “Harmony Behavioral Healthcare” so as to “put some distance between BH [Harmony] and the WellCare name.” Farha knew that the success of the Harmony scheme depended upon keeping a low profile and avoiding an audit. The evidence also shows that subordinates at WellCare routinely apprised Farha of the 80/20 reporting process. Farha knew generally when the 80/20 Worksheets arrived. He knew which employees were taking charge of the reports. A steady stream of emails kept Farha informed, from which a jury could reasonably infer Farha’s active oversight and coordination. In 2004—the year in which Staywell and HealthEase submitted their CY 2002 and 2003 expense reports—Kale regularly emailed Farha detailed updates regarding Harmony, some of which concerned WellCare’s strategies in addressing the 80/20 rule. Farha was frequently in touch with Bereday as well. Bereday later emailed Farha requesting clearance for Staywell and HealthEase to submit their finalized 80/20 expense reports to AHCA based on the calculations produced by the Medical Economics team. Farha gave clearance and signed the accompanying certifications. 86 Case: 14-12373 Date Filed: 08/11/2016 Page: 87 of 124 Farha stayed involved in subsequent years. For the CY 2004 reporting cycle, after the 80/20 Worksheets and cover letters arrived from AHCA, Farha sent an email to Behrens and Kale, among others, saying, “Team, lets [sic] be sure we handle this one appropriately. Who is on point for this process?” Behrens responded that he was, along with his team. At one point, Farha and Bereday discussed a slide presentation relating to the 80/20 rule. The presentation showed both the expenses Staywell and HealthEase had submitted to AHCA in their CY 2004 reports and their much lower actual qualifying expenses. Farha’s supervision continued during the CY 2005 reporting cycle. For example, Farha was privy to an email exchange between Staywell and HealthEase president Imtiaz Sattaur and Behrens in which Sattaur explained, “[T]he plan is that we stay consistent to last year’s reporting by utilizing our Harmony BH Sub methodology, less inpatient costs. We will review the final report with Todd before we send it to AHCA.” And they did. Before finalizing the 80/20 figures for the CY 2005 expense reports, Behrens slipped into Farha’s office to confirm that “1.4 is okay.” Staywell and HealthEase collectively refunded a total of $1.4 million to AHCA for CY 2005. The $1.4 million figure was a fabricated and false number, which Farha knew. And it was Farha who gave the annual fraudulent refund targets. For CY 2002 and 2003, Farha told his subordinates “to find a way not to pay back 10 87 Case: 14-12373 Date Filed: 08/11/2016 Page: 88 of 124 million dollars” as WellCare’s initial refund forecast had projected, but instead, to “find[] a way to make it zero.” By the time of the CY 2005 reporting year, the target had moved. Though Clay proposed a methodology that would result in no refund, Farha insisted on a different reporting strategy, ordering, “No, we’re not going to do it like that. You have to pay the Gods something.” Instead, they would “pay back a million.” A reasonable jury could view Farha’s order as evidence that Farha wanted Staywell and HealthEase to refund just enough to avoid scrutiny, thereby protecting WellCare’s large ill-gotten profits. 19 Farha’s repeated refusals to allow those at WellCare to disclose to AHCA that Staywell and HealthEase were reporting sub-capitation payments to Harmony (rather than reporting what they paid providers for CMH/TCM services) were additional evidence from which a jury could infer Farha’s fraudulent intent. Farha participated in efforts by other industry players and the Florida Association of Health Plans to negotiate 80/20-eligible expenses with AHCA. On multiple occasions throughout this process, Sattaur urged Farha to disclose to AHCA that 19 Farha offers another take on these statements, arguing that his order to pay some refund amount rather than no refund should be construed as evidence of prudence and conservativism motivated by legitimate business reasons rather than evidence of fraudulent intent. But the jury was free to draw different conclusions regarding Farha’s true motives. The context in which Farha gave refund targets permits an inference of intent to defraud. Farha’s predetermined refund targets were inconsistent with Staywell’s and HealthEase’s obligations to report actual CMH/TCM expenses and refund the difference between actual expenses and 80%. A jury could reasonably infer from Farha’s ordered predetermined refund targets that Farha knew the expense figures in the 80/20 expense reports would, in fact, be false. Vernon, 723 F.3d at 1273. 88 Case: 14-12373 Date Filed: 08/11/2016 Page: 89 of 124 Staywell and HealthEase had been reporting sub-capitation payments to Harmony since WellCare had not revealed this fact to AHCA. Each time, Farha declined to do so. Sattaur testified that Farha was confident that through lobbying efforts and his ability to influence the Secretary of AHCA, the 80/20 law would soon be repealed and that the issue would blow over. The evidence established Farha also made sure his subordinates did not disclose Staywell’s and HealthEase’s reporting practices to AHCA either. At one point, Farha attended an 80/20-related company meeting regarding the negotiations with AHCA. At that meeting, Michael Turrell, a WellCare lawyer who worked under Bereday, was told not to disclose to AHCA or other industry players details that would reveal how Staywell and HealthEase calculated their 80/20 expenses. Turrell reassured Farha that he had appropriately screened his comments when communicating with other parties. West similarly testified that both Behrens and Bereday told him, on different occasions, to not call AHCA. These exchanges are evidence from which a reasonable jury could infer a collective policy of secrecy on the part of WellCare’s leadership. Farha’s insistence on secrecy was evidence from which a reasonable jury could infer fraudulent intent. 20 20 Defendants emphasize that AHCA approved Staywell’s and HealthEase’s subcontracts with Harmony. But the Harmony subcontracts do not show, or even suggest, that AHCA knew that Staywell’s and HealthEase’s reported expenses were manipulated and fabricated figures and not what they had actually paid providers of CMH/TCM services. In his reply brief, Farha takes a different tact, arguing that he had legitimate strategic reasons for not wanting AHCA to discover Staywell and HealthEase’s reporting methodology. 89 Case: 14-12373 Date Filed: 08/11/2016 Page: 90 of 124 From Farha’s exchanges with his subordinates, a reasonable jury could also infer that Farha continued to be actively involved in overseeing and directing the 80/20 reporting process. Farha’s subordinates routinely checked in with him, provided him with updates, and received orders about the size of the refund Staywell and HealthEase were to remit to AHCA. All these communications confirmed that Behrens’s team continued to prepare and submit the 80/20 expense reports consistent with Farha’s scheme. There was no need for Farha to micromanage the 80/20 reporting once he designed the scheme, worked out the logistics, and delegated the pertinent tasks. By the CY 2006 reporting cycle, Behrens’s Medical Economics team handled the particulars in preparing Staywell’s and HealthEase’s expense reports, and the emails they circulated among themselves did not include Farha.21 Nonetheless, Farha ignores that for CY 2006, he signed a WellCare policy and procedure document, as he had done before, acknowledging Staywell’s and HealthEase’s statutory and contractual duties to comply with the 80/20 He argues “it would have signaled that WellCare viewed the BHO question as an open one.” Farha concludes that “[n]o negative inference can fairly be drawn from the decision not to invite senior AHCA officials to treat the issue as a subject for negotiation.” Given all the evidence, however, the jury was not required to accept Farha’s argument. The jury was free instead to infer that Farha’s posture of secrecy and nondisclosure to AHCA was part of his fraudulent reporting scheme. 21 Farha was shrewd about paper trails. For example, in 2006, after one of Farha’s subordinates sent a lengthy email to a number of WellCare personnel about its strategy in engaging with AHCA regarding the 80/20 rule, Farha wrote back, “[T]his is too large a distribution for anything confidential. Sensitive items are best handled verbally with those who must know.” 90 Case: 14-12373 Date Filed: 08/11/2016 Page: 91 of 124 requirements. Farha even signed the refund checks Staywell and HealthEase issued to AHCA in conjunction with submitting their CY 2006 80/20 expense reports. Contrary to his contentions, Farha did more than just devise a scheme to defraud AHCA or commit a mere act in furtherance of executing that scheme. Farha was CEO, President, and a director of WellCare. As such, he not only devised, but implemented and supervised the scheme’s execution year after year. In fact, Sattaur provided a summary of Farha’s role in the scheme to defraud AHCA. He explained that after Clay and the Medical Economics team had calculated Staywell’s and HealthEase’s reported expenses, and after Behrens had approved their work, “the ultimate sign-off on the approval of whether [an 80/20 report] gets filed with the State would be by Mr. Todd Farha.” Sattaur testified that Farha, Behrens, and Bereday together were “in charge” of WellCare’s policy of using the fraudulent reporting method concerning “whether it [was] the right thing to do.” Sattaur repeatedly urged Farha to disclose to AHCA that Staywell and HealthEase had reported what they paid Harmony (rather than what they paid providers through Harmony), but Farha refused. Sattaur explained that he himself never considered disclosing this fact to AHCA because the decision of what to disclose to AHCA “was being worked by Mr. Todd Farha and his team of government affairs.” Sattaur explained that “there was a very tight control over 91 Case: 14-12373 Date Filed: 08/11/2016 Page: 92 of 124 that issue with Todd Farha and his team that if you were to break the plan that they [had], that would not be a good thing to do.” It “could be tantamount to jeopardizing your career at WellCare.” In summary, the evidence sufficiently showed that Farha aided and abetted the execution of the fraud in the year for which he was convicted, and he did so knowingly, willfully, and with intent to defraud AHCA. Accordingly, the evidence was sufficient to sustain his convictions for Medicaid fraud, in violation of 18 U.S.C. § 1347. G. Advice of Counsel Evidence Defendants point to communications and testimony by lawyers who worked for WellCare to claim the defendants were told that their CMH/TCM reporting method was legal and common practice in the industry. The evidence showed outside counsel contacted Florida Health Partners (“FHP”) and learned FHP sub-capitated to “related entities,” and this “seemed” acceptable “under the 80/20 calculation to AHCA.” The defendants concede, however, that the “related entities” to which FHP made sub-capitated payments were actual health clinics that provided medical services. If anything, this showed the defendants that they should not count money paid to a related company that, like Harmony, provided no health care services to Medicaid patients. 92 Case: 14-12373 Date Filed: 08/11/2016 Page: 93 of 124 Outside counsel also learned that United Health Plans (“United”) “used” payments it made to a “related specialty organization” United Behavioral Health “in connection with the 80/20 calculation.” The defendants ignore that outside counsel, when reporting to general counsel Bereday, said that though United “did it in this certain fashion . . . the mere fact that” it did so “doesn’t necessarily mean that that method is or will be approved by AHCA now or in the future.” More importantly, outside counsel was asked to “render a clean opinion” concerning “use of . . . all of the contract expenses between [Staywell and HealthEase] and Harmony for purposes of meeting the 80/20 requirement.” Outside counsel was unwilling to give a “clean opinion,” that is, “a legal opinion that in all probability would be upheld if there were any kind of problems or allegations or appeals.” WellCare’s former outside counsel testified that, after refusing to give a clean opinion as to the Harmony reporting method, “the number of assignments and the . . . work referred to us by the client diminished dramatically.” Outside counsel testified he told those at WellCare that, if Staywell and HealthEase were going to use the Harmony reporting method, “they should (a) tell the agency about it and (b), more importantly, make a rule challenge or declaratory judgment action, some action to put these disputed . . . policy issues in front of an impartial officer.” 93 Case: 14-12373 Date Filed: 08/11/2016 Page: 94 of 124 In the light most favorable to the jury’s verdict, this advice-of-counsel evidence hurts, not helps, the defendants. If anything, outside counsel’s advice warned the defendants not to use their Harmony reporting method without informing AHCA. The defendants, however, proceeded in secrecy. This evidence does not undermine the jury verdict given the abundant evidence of the defendants’ intent to defraud AHCA. H. Clay’s § 1001 False Statements Clay challenges his two convictions in Counts 10 and 11 for making false statements to federal agents, in violation of 18 U.S.C. § 1001. To convict Clay under § 1001, the government had to prove “(1) that a statement was made; (2) that it was false; (3) that it was material; (4) that it was made with specific intent; and (5) that it was within the jurisdiction of an agency of the United States.” United States v. House, 684 F.3d 1173, 1203 (11th Cir. 2012). Clay argues the government failed to present sufficient evidence of: (1) falsity, (2) willfulness, and (3) materiality. Count 10 of the indictment charged that Clay told federal agents that Staywell and HealthEase had not over-reported outpatient behavioral health care expenses to AHCA to reduce the refunds paid to AHCA, when in fact, Clay knew that the expense figures in the CY 2005 Worksheets were purposefully over- reported to reduce refunds paid to AHCA. Count 11 charged that Clay told federal 94 Case: 14-12373 Date Filed: 08/11/2016 Page: 95 of 124 agents that Staywell and HealthEase had not purposefully inflated the costs associated with their behavioral health care encounter data submissions to AHCA, when in fact, Clay knew Staywell and HealthEase had done so in February 2007. We consider the sufficiency of the evidence for Clay’s § 1001 convictions. 1. Falsity Clay’s statements to the federal agents were proven false. 22 Agent Vic Milanes asked Clay if Staywell and HealthEase had over-reported their outpatient behavioral health costs to AHCA over the years in order to avoid paying money back to AHCA. Clay responded that, to his knowledge, they had not. But Clay knew the opposite was true. Clay worked with West and others on Behrens’s team to produce the CY 2005 expense reports. West conferred with Clay in producing the calculations for the reports, West reported to Clay the results of his work, and Clay was among those present in Bereday’s office the day West presented his work and WellCare certified the CY 2005 reports. It was Clay who relayed to West that “Farha wants to pay back a million” because “[y]ou have to pay the Gods something.” Clay 22 As to the specific § 1001 charges in Counts 10 and 11, the district court instructed the jury that the government had to prove: (1) “the defendant made a statement as charged”; (2) “the statement was false”; (3) “the falsity concerned a material matter”; (4) “the defendant acted willfully knowing that the statement was false”; and (5) “the false statement was made or used for a matter within the jurisdiction of the department or agency of the United States.” The court instructed that “[a] statement is false when made if it is untrue when made and the person making it knows it is untrue.” On appeal, Clay does not challenge the court’s charge as to § 1001. To the extent Clay challenges the general part of the jury charge, his claims lack merit. 95 Case: 14-12373 Date Filed: 08/11/2016 Page: 96 of 124 knew Staywell and HealthEase had used both double-counting and premium- difference calculations in an effort to achieve Farha’s desired result. It was Clay’s idea to use the premium difference calculation in the first place. Clay knew that for CY 2005, Staywell and HealthEase combined had reported $18,462,421 in 80/20 expenses and had refunded only $1,440,449 to AHCA. Clay also knew that the internal spreadsheets, prepared by West, showed that Staywell and HealthEase combined had actually paid health care providers only $12,956,122 for qualifying services. The 80/20 Worksheets and cover letters instructed Staywell and HealthEase to report money paid to providers of CMH/TCM services. Clay knew that if Staywell and HealthEase reported $12 million in CMH/TCM expenses, they would owe a refund of $6,946,748. Clay admitted as much in an email he sent to Behrens just days before the CY 2005 expense reports were certified, stating, “If we took AHCA payments and AHCA definitions of eligible care we would owe them $6.9 million.” In Clay’s own words, “the whole reason Harmony exist[ed]” was to “hide” the “big slug” of profits that Harmony captured for WellCare. Clay summed up Staywell’s and HealthEase’s approach to 80/20 compliance this way: “Every year we’ve fed the gods. We’ve paid them a little money to keep them happy. We’ve paid them a million bucks a year, or whatever.” Clay knew Staywell and 96 Case: 14-12373 Date Filed: 08/11/2016 Page: 97 of 124 HealthEase had over-reported their 80/20 expenses, but he falsely told Agent Milanes that they had not. Agent Milanes also asked Clay whether Staywell and HealthEase had purposefully inflated the costs of their behavioral health encounter submissions to AHCA. Clay responded that, to his knowledge, they had not. But, again, Clay knew the opposite was true. Clay attended secretly-recorded company meetings on both January 16, 2007, and January 29, 2007. At the first meeting, convened to discuss how Staywell and HealthEase planned to price their patient-provider encounters, Clay listened as both Robert Butler and Marc Ryan suggested that Staywell and HealthEase price the encounters based on what Harmony paid health care providers. Clay heard Butler explain that encounter prices should reflect only actual costs in money paid to providers, rather than administration, overhead, and profit for Harmony, because AHCA already built into the capitated rate money for administration, overhead, and profit. Clay, however, warned that if they did what Butler said they should do, they would be in “deep trouble.” Clay offered his take on the encounter data reporting process: “[T]he state is doin’ this as another end around, to find out how much money we’re makin’ in that. Which we’ve been finessing, for years.” If they priced encounters based on what Harmony paid providers, Clay knew it would 97 Case: 14-12373 Date Filed: 08/11/2016 Page: 98 of 124 reveal to AHCA their huge, ill-gotten profits, and Clay feared there would be a “massive rate cut” in Staywell’s and HealthEase’s premium money. At one point Clay explained to his colleagues, “[I]f you price [the encounter data] at anything reasonable, we’re gonna show a 50% loss ratio, and we’re right back to opening the Kimono.” As a solution, Clay proposed they instead price the encounters based upon the sub-capitation payments to Harmony. This method spread the full sub- capitation sum across all of the encounters and resulted in encounter prices that were significantly higher than what Harmony, had paid providers of CMH/TCM services. In the January 29 meeting, Clay told his team, “I think we’re going to have to put some numbers that are about 40 percent higher than we think they should be, because we’re making about a 40 percent profit margin. And that’s what we’re gonna submit . . . . And that’s all there is to this conversation. It’s that simple.” He later added, “[I]t’s just a matter of how inflated a unit cost number we’re going to be submitting.” Clay knew Staywell and HealthEase had purposefully inflated the costs of their behavioral health encounter data. The jury’s task was to determine what Clay understood the agents to be asking him and whether Clay knew what he told the federal agents was false. The 98 Case: 14-12373 Date Filed: 08/11/2016 Page: 99 of 124 evidence was more than adequate for the jury to find Clay understood the question and knew his answer was false.23 It is noteworthy that Staywell’s and HealthEase’s CY 2005 expense reports for CMH/TCM services contained expense figures calculated using both their sub- capitation payments to Harmony and Harmony’s payments to providers, a double- counting calculation method. Even if Clay thought the sub-capitation payments themselves were a reportable expense (which he did not), he still knew the reported expenses in CY 2005 involved double-counting and were therefore “over- reported.” For example, West’s spreadsheets reveal that for CY 2005, Staywell and HealthEase collectively paid Harmony $13,507,701, and Harmony paid providers $12,956,122. In 2006, WellCare executives realized they forgot to update the Harmony subcontracts so as to pay Harmony any of the additional premium money Staywell and HealthEase received for the CMH/TCM program expansion. The $13,507,701 they did pass along to Harmony was enough to account for the $12,956,122 that Harmony paid providers. But in addition to reporting expenses of $13,507,701 to AHCA, Staywell and HealthEase double-counted a portion of the 23 Clay invokes the Whiteside decision, but stretches it to mean that the agent’s factual questions to him were necessarily posed and answered under Clay’s interpretation of the 80/20 rule. Clay argues his denials were true based on his reasonable legal interpretation of the questions asked, but his interpretation is not reasonable. We reject Clay’s Whiteside arguments for the same reasons outlined above as to the other defendants, including the fact that the evidence showed Clay did not believe his post-hoc interpretation anyway. 99 Case: 14-12373 Date Filed: 08/11/2016 Page: 100 of 124 $12,956,122 sum, which was already accounted for in the $13,507,701 figure. Kelly, the forensic accountant, explained why there was no valid basis for such an accounting maneuver. Clay knew the 80/20 expenses were over-reported. As to encounter data, Clay quarrels with the meaning of the word “inflated” in Agent Milanes’s question. The jury heard that as of the October 24, 2007 raid on WellCare, federal investigators had already gathered months’ worth of secretly- recorded company conversations collected by a whistleblower. Though Clay did not necessarily know that, Clay knew that investigators were executing a search warrant of WellCare’s corporate headquarters. Clay knew, based on the questions the agents asked him, that the search concerned WellCare’s 80/20 expense and encounter data reporting. Agent Milanes’s questions concerned, in part, what had been discussed at WellCare company meetings that Clay attended. When Agent Milanes asked Clay whether Staywell and HealthEase had purposefully “inflated” their encounter costs, Agent Milanes knew that Clay had previously told his colleagues, “I think we’re going to have to put some numbers that are about 40 percent higher than we think they should be . . . . [I]t’s just a matter of how inflated a unit cost number we’re going to be submitting.” (emphasis added). 24 24 Clay cites to the encounter template in the contract, which provided that Staywell and HealthEase were to report the “[a]mount [p]aid” per encounter as the “[c]osts associated with the claim.” Clay argues that the “[c]osts associated with the claim” reasonably could be interpreted to be either (1) what Harmony actually paid the provider for the claim or (2) an allocable (larger) portion of what Staywell or HealthEase paid Harmony to cover the claim, which included Harmony’s administration, overhead, and profit generated for WellCare. This is not a reasonable 100 Case: 14-12373 Date Filed: 08/11/2016 Page: 101 of 124 Clay had no basis to deny that Staywell and HealthEase had purposefully inflated their encounter costs. Clay was, of course, free to argue to the jury that he understood Agent Milanes to be asking something else, but the duties of fact finding and making credibility determinations belong to the jury alone.25 Our role is simply to ensure there was sufficient evidence from which a reasonable jury could find that Clay understood what Agent Milanes was asking and that Clay knew his denials were false. That evidence was amply sufficient. 2. Willfulness Clay’s statements to the agents were not only false but willfully made. As to specific intent, § 1001 criminalizes false statements made “knowingly and willfully.” 18 U.S.C. § 1001; see also House, 684 F.3d at 1204. The Supreme Court has said that, to establish a willful violation of a statute, generally “the Government must prove that the defendant acted with knowledge that his conduct was unlawful.” Bryan v. United States, 524 U.S. 184, 191-92, 118 S. Ct. 1939, 1945 (1998). Using the pattern instruction, the district court charged the jury that “[t]he word ‘willfully’ means that the act was committed voluntarily and purposefully with the intent to do something the law forbids, that is, with the bad interpretation of the federal agent’s question. In any event, tape recordings of Clay established that Clay knew the encounter data was inflated. 25 Agent Johnston testified that he did not recall Clay asking for clarification of any questions that Agent Milanes asked. 101 Case: 14-12373 Date Filed: 08/11/2016 Page: 102 of 124 purpose to disobey or disregard the law.” See Eleventh Circuit Pattern Jury Instructions (Criminal Cases) 9.1A (2010). Here, the context of Clay’s statements is important. Clay met with Agents Johnston and Milanes on the same day that over 200 federal agents streamed through the doors of WellCare’s corporate office to execute a search warrant. Agent Johnston testified that when they approached Clay to ask if he would consent to an interview, Johnston and Milanes would have identified themselves as federal agents involved with the execution of the search warrant. Johnston testified that they would have shown Clay their federal credentials and that Clay knew he and Milanes were federal agents. Johnston testified that he and Milanes also told Clay they wanted to interview him in conjunction with the investigation of WellCare. The agents proceeded to interview Clay in his office for approximately an hour and a half. During this time, other agents continued executing the search warrant outside Clay’s office. Many of the interview questions Agent Milanes asked Clay specifically pertained to WellCare’s reporting of the 80/20 expenses and encounter data. Clay served as Vice President of Medical Economics under Behrens. He was closely involved with the preparation of Staywell’s and HealthEase’s CY 2005 expense reports for CMH/TCM services and their February 2007 encounter data submissions. He fully knew about their reporting obligations to AHCA. Given 102 Case: 14-12373 Date Filed: 08/11/2016 Page: 103 of 124 Clay’s job and the subject of Agent Milanes’s questions, the jury could readily infer that Clay knew the agents were investigating WellCare’s reporting of its actual expenses to AHCA. By knowingly making false statements to the federal agents during the raid, Clay acted willfully. See Bryan, 524 U.S. at 191-92, 118 S. Ct. at 1945. Clay argues that the government was required to offer more mens rea evidence of willfulness. To be sure, the government’s evidence of willfulness was circumstantial, but “[g]uilty knowledge can rarely be established by direct evidence, especially in respect to fraud crimes which, by their very nature, often yield little in the way of direct proof.” United States v. Suba, 132 F.3d 662, 673 (11th Cir. 1998). Mens rea elements such as knowledge or intent may be proven by circumstantial evidence. See United States v. Santos, 553 U.S. 507, 521, 128 S. Ct. 2020, 2029 (2008); Suba, 132 F.3d at 673. The government did not need to rely, and did not rely, on a presumption of willfulness to prove Clay violated § 1001. The government presented ample evidence from which a reasonable jury could infer that Clay acted willfully and with the necessary criminal intent. 3. Materiality As to materiality, Clay’s false statements concerned the core conduct that the agents were investigating during the October 2007 raid of WellCare. Clay’s 103 Case: 14-12373 Date Filed: 08/11/2016 Page: 104 of 124 false denials of over-reporting and inflating encounter prices went to the heart of the matter being investigated and were material. Contrary to Clay’s contention, the test is not whether the agents were actually misled. A false statement is material if it has “a natural tendency to influence, or [is] capable of influencing, the decision of the decisionmaking body to which it was addressed.” United States v. Boffil-Rivera, 607 F.3d 736, 741 (11th Cir. 2010). “[A] false statement can be material even if the decision maker actually knew or should have known that the statement was false” or “even if the decision maker did not actually rely on the statement.” United States v. Neder, 197 F.3d F.3d 1122, 1128 (11th Cir. 1999); see also House, 684 F.3d at 1203 (“[P]roof of actual influence is not required.”); United States v. Gafyczk, 847 F.2d 685, 691 (11th Cir. 1988) (“[I]n order for a false statement to be material under § 1001, it need not be shown to have actually influenced the government or caused it any pecuniary loss.”). Clay was aware he was being interviewed by federal investigators and that WellCare was being investigated precisely for the exact conduct he was being asked about. Clay was a high-level, sophisticated executive at a publicly-traded company receiving public funds, and he certainly knew that lying to federal agents investigating the company for health care fraud was unlawful. He was undoubtedly familiar with the dozens of certifications and warnings that false 104 Case: 14-12373 Date Filed: 08/11/2016 Page: 105 of 124 statements to the government carry criminal liability. Nevertheless, he told the agents there was no over-reporting and no inflation, despite knowing these things were not true. Ample evidence allowed a reasonable jury to find Clay knowingly and willfully made false material statements to federal agents. VI. JURY INSTRUCTIONS Farha, Behrens, and Kale challenge their fraud convictions under 18 U.S.C. § 1347 and contend the district court improperly instructed the jury as to their knowledge that the reported expenses were false. Regarding the defendants’ § 1347 charges, the district court instructed the jury that the defendants had to act knowingly, willfully, and with intent to defraud: A defendant can be found guilty of this offense only if all the following facts are proved beyond a reasonable doubt: One, he knowingly executed or attempted to execute a scheme or artifice to defraud a healthcare benefit program or to obtain money or property owned by or under the custody or control of a healthcare benefit program by means of false or fraudulent pretenses and representations; [T]wo, the false or fraudulent pretenses and representations related to a material fact; [T]hree, he acted willfully and intended to defraud; and [F]our, he did so in connection with the delivery of or payment for healthcare benefits, items, or services. 105 Case: 14-12373 Date Filed: 08/11/2016 Page: 106 of 124 (emphasis added). The district court explained that “knowingly” means that “an act was done voluntarily and intentionally and not because of a mistake or by accident.” The court charged that “willfully” means that “the act was committed voluntarily and purposely with the intent to do something the law forbids, that is, with the bad purpose to disobey or disregard the law.” The instructions thus advised the jury that it could not find the defendants guilty unless it concluded that they acted voluntarily, intentionally and with the bad purpose to disregard the law in executing a scheme to defraud AHCA. The district court also instructed that “[a] scheme to defraud includes any plan or course of action intended to deceive or cheat someone out of money or property by using false or fraudulent pretenses and representations relating to a material fact.” The district court then instructed that a “statement or representation is false or fraudulent if it is about a material fact that the speaker knows is untrue or makes with deliberate indifference as to the truth and makes with intent to defraud.” The court added, “A statement or representation may be false or fraudulent when it’s a half truth or effectively conceals a material fact and is made with the intent to defraud.” The court explained that “‘[t]o act with intent to defraud’ means to do something with a specific intent to deceive or cheat someone and to deprive someone of money or property.” 106 Case: 14-12373 Date Filed: 08/11/2016 Page: 107 of 124 The district court thus told the jury that, as to knowledge of falsity, the defendants had to either “know” the representations were untrue or make them “with deliberate indifference as to the truth” and “with intent to defraud.” This deliberate indifference instruction was tethered to an instruction requiring a finding that the defendants made the representations “with intent to defraud.” A. Standard of Review We review jury instructions “to determine whether the instructions misstated the law or misled the jury to the prejudice of the objecting party.” United States v. Gibson, 708 F.3d 1256, 1275 (11th Cir. 2013). We will not reverse a conviction based on a jury instruction challenge “unless we are ‘left with a substantial and ineradicable doubt as to whether the jury was properly guided in its deliberations.’” Id. But “[w]hen the jury instructions, taken together, accurately express the law applicable to the case without confusing or prejudicing the jury, there is no reason for reversal even though isolated clauses may, in fact, be confusing, technically imperfect, or otherwise subject to criticism.” Id. Moreover, the Supreme Court has admonished that “in reviewing jury instructions, our task is also to view the charge itself as part of the whole trial,” noting that “[o]ften isolated statements taken from the charge, seemingly prejudicial on their face, are not so when 107 Case: 14-12373 Date Filed: 08/11/2016 Page: 108 of 124 considered in the context of the entire record of the trial.” United States v. Park, 421 U.S. 658, 675-76, 95 S. Ct. 1903, 1913 (1975). 26 B. Section 1347 Instructions The defendants argue that the district court erred by instructing the jury that it could convict the defendants under § 1347 upon finding that the defendants made false representations in the CY 2006 expense reports “with deliberate indifference as to the truth.” They argue that the “deliberate indifference” standard is akin to a “recklessness” standard and impermissibly lowered the bar below what Vernon and Medina require. We agree that in a health care fraud case such as this, “the defendant must be shown to have known that the claims submitted were, in fact, false.” United States v. Vernon, 723 F.3d 1234, 1273 (11th Cir. 2013) (quoting United States v. Medina, 485 F.3d 1291, 1297 (11th Cir. 2007)). Although the government must prove the defendant’s knowledge of falsity, a defendant’s knowledge can be proven in more than one way. Here, the district court properly instructed the jury that a “statement or representation is false or fraudulent if it is about a material fact that the speaker knows is untrue or makes with deliberate indifference as to the truth and makes 26 We review de novo the legal correctness of jury instructions, but we review the district court’s phrasing for abuse of discretion. United States v. Prather, 205 F.3d 1265, 1270 (11th Cir. 2000). Jury instructions are also subject to harmless error review. United States v. House, 684 F.3d 1173, 1196 (11th Cir. 2012). An error is harmless “if the reviewing court is satisfied ‘beyond a reasonable doubt that the error complained of did not contribute to the verdict obtained.’” Id. at 1197. 108 Case: 14-12373 Date Filed: 08/11/2016 Page: 109 of 124 with intent to defraud.” Representations made with deliberate indifference to the truth and with intent to defraud adequately satisfy the knowledge requirement in § 1347 cases. The Eleventh Circuit Pattern Jury Instructions for § 1347 support our conclusion. The pattern § 1347 instruction provides that a “statement or representation is ‘false’ or ‘fraudulent’ if it is about a material fact that the speaker knows is untrue or makes with reckless indifference as to the truth and makes with intent to defraud.” Eleventh Circuit Pattern Jury Instructions (Criminal Cases) 53 (2010). Here, the district court’s instruction mirrored the § 1347 pattern instruction, except the district court used the even stronger phrase “deliberate indifference” instead of the phrase “reckless indifference” found in the pattern instructions. Id. The district court’s language imposed a higher burden on the government than that suggested by our § 1347 pattern jury instruction. In the mail and wire fraud context, this Court has said that “[f]raudulent conduct that will establish a ‘scheme to defraud’ includes knowingly making false representations” and also “statements made with reckless indifference to their truth or falsity.” United States v. Sawyer, 799 F.2d 1494, 1502 (11th Cir. 1986); see also United States v. Simon, 839 F.2d 1461, 1470 (11th Cir. 1988) (“[R]eckless indifference to the truth . . . supplies the criminal intent necessary to convict . . . .”); United States v. Edwards, 458 F.2d 875, 881 (11th Cir. 1972) (“Such reckless 109 Case: 14-12373 Date Filed: 08/11/2016 Page: 110 of 124 indifference to the truth of representations is more than sufficient to afford the government a remedy under the mail fraud statute.”). The district court’s instruction was not only consistent with the pattern charge but also with this Circuit’s fraud precedents. The defendants argue the mail and wire fraud cases are inapplicable because those statutes do not require, as § 1347 does, that a defendant “knowingly and willfully execute[] . . . a scheme to defraud.” 18 U.S.C. § 1347(a); see 18 U.S.C. §§ 1341, 1343. But § 1347 links knowledge and willfulness to a “scheme to defraud.” See 18 U.S.C. § 1347(a). The government thus had to prove the defendants both (1) knowingly and willfully executed that scheme to defraud and (2) made false statements with “deliberate indifference as to the truth” and “with intent to defraud.” Cf. United States v. Dearing, 504 F.3d 897, 903 (9th Cir. 2007) (holding that where a court’s “reckless indifference” instruction “was tethered to the specific intent to defraud element” of § 1347, such an instruction did not negate the court’s separate instruction that, to convict under § 1347, the jury also had to find the defendant “knowingly and willingly” executed a scheme to defraud). Here, the district court properly defined “knowingly” and “willfully” and made clear the government had to prove that the defendants executed a scheme to defraud AHCA “voluntarily and intentionally” rather than by “mistake or by accident” and “with the intent to do something the law forbids.” The district court 110 Case: 14-12373 Date Filed: 08/11/2016 Page: 111 of 124 also linked “deliberate indifference” to “intent to defraud.” The instruction required the jury to find more than deliberate indifference to the truth; rather, a finding of deliberate indifference would suffice only if the jury also found that the defendants made the false statement with intent to defraud. The court then instructed that “‘[t]o act with intent to defraud’ means to do something with a specific intent to deceive or cheat someone and to deprive someone of money or property.” Therefore, under the factual circumstances of this case, to find that the defendants made representations of expenses in the CY 2006 reports (1) with deliberate indifference to the truth and (2) with intent to defraud necessarily required the jury to find that the defendants knew the representations were false. See United States v. Hough, 803 F.3d 1181, 1197-98 (11th Cir. 2015) (“A crucial assumption underlying the jury trial system is that juries will follow the instructions given them by the trial judge.” (alterations omitted)); United States v. Stone, 9 F.3d 934, 938 (11th Cir. 1993) (“Few tenets are more fundamental to our jury trial system than the presumption that juries obey the court’s instructions.”). Further, the trial proceeded under a theory of actual knowledge rather than deliberate indifference. The indictment charged that the defendants knew the information in the CY 2006 reports was false. The government’s closing argument hammered over and over again that the defendants knew what they represented to AHCA was false. From beginning to end, the government alleged the defendants’ 111 Case: 14-12373 Date Filed: 08/11/2016 Page: 112 of 124 knowledge and intent, not mere recklessness. For the jury to convict the defendants without finding that they knew the expense reports were false would be to ignore both the district court’s jury instructions and the government’s whole theory of the case. Viewing the charge as a whole and the entire trial, we find no error, much less reversible error, in the court’s thorough charge to the jury. C. Willful Blindness Instruction The defendants also argue that our circuit’s pattern § 1347 instruction is inconsistent with the Supreme Court’s decision in Global-Tech Appliances, Inc. v. SEB S.A., 563 U.S. 754, 131 S. Ct. 2060 (2011). They argue that, under Global- Tech, the district court should have instructed that the government had to prove (1) actual knowledge of falsity or (2) at least “willful blindness.” Id. at 769; 131 S. Ct. at 2070. First, as the government emphasizes, the district court actually did give a “willful blindness” instruction consistent with the definition of willful blindness in Global-Tech. The Supreme Court in Global-Tech said that willful blindness has “two basic requirements: (1) the defendant must subjectively believe that there is a high probability that a fact exists and (2) the defendant must take deliberate actions to avoid learning of that fact.” Id. at 769, 131 S. Ct. at 2071. In the opening part of its charge, here, the district court similarly instructed that “[i]f a defendant’s knowledge of a fact is an essential part of a crime, it is enough that the defendant 112 Case: 14-12373 Date Filed: 08/11/2016 Page: 113 of 124 was aware of a high probability that the fact existed and took deliberate action to avoid learning of the fact unless the defendant actually believed the fact did not exist.” The district court gave an example to explain “deliberate” action to avoid knowledge of a fact: To give you an example from a different kind of case, deliberate avoidance of positive knowledge, which is equivalent of knowledge, occurs in a drug case if a defendant possesses a package and believes it contains a controlled substance but deliberately avoids learning that it contains the controlled substance so he or she can deny knowledge of the package’s contents. So, in such a case, the jury may find that a defendant knew about the possession of a controlled substance if the jury determines beyond a reasonable doubt that the defendant, one, actually knew about the controlled substance or, two, had every reason to know but deliberately closes his or her eyes. (emphasis added). The court then admonished: “But I must emphasize that negligence, recklessness, carelessness, or foolishness is not enough to prove that a defendant knew about the possession of the controlled substance.” (emphasis added). Alternatively, the defendants argue that the district court still erred by allowing the jury to find knowledge of falsity under the § 1347 pattern instruction standard of “reckless indifference to the truth and intent to defraud” as opposed to charging only the Global-Tech standard of actual knowledge or willful blindness.27 27 Behrens (as to § 1035) and Clay (as to § 1001) also argue the evidence was insufficient to trigger a willful blindness instruction at all. For example, Behrens’s brief (and Clay’s by 113 Case: 14-12373 Date Filed: 08/11/2016 Page: 114 of 124 This ignores that the district court substituted “deliberate indifference” for “reckless indifference” in the § 1347 pattern charge. That substitution made the § 1347 pattern charge much closer to the “deliberate” standard in the willful blindness charge. The district court never once said “reckless indifference.” The court explicitly said that recklessness was not enough. We also reject the claim that Global-Tech alone controls this criminal § 1347 fraud case or creates reversible error here. Global-Tech is a civil patent- infringement case. In Global-Tech, the Supreme Court analyzed the meaning of the term “actively induces” in 35 U.S.C. § 271(b), a statute that provides that “[w]hoever actively induces infringement of a patent shall be liable as an infringer.” Id. 563 U.S. at 760, 131 S. Ct. at 2065. Global-Tech was not a criminal fraud case and did not abrogate, conflict with, or preclude the district court from giving the § 1347 pattern charge in this case. As noted earlier, knowledge of falsity can be proved in more than one way, and we view the § 1347 pattern charge adoption) argues that there was no evidence the defendants were aware of a high probability that their expense reports were false and purposefully contrived to avoid learning the truth. We disagree and find adequate evidence to warrant the instruction, given their deliberate refusal to call AHCA at certain important times. Although there was more evidence of actual knowledge, we cannot say the evidence of willful blindness was non-existent or too sparse. Alternatively, given the abundant evidence of actual knowledge, any alleged error was harmless for the reasons outlined in United States v. Esquenazi, 752 F.3d 912, 931 (11th Cir. 2014), and Stone, 9 F.3d at 938-39. 114 Case: 14-12373 Date Filed: 08/11/2016 Page: 115 of 124 as a permissible and acceptable way to prove knowledge of falsity. 28 Considering the charge as a whole, we conclude that the district court did not err in giving the § 1347 charge in this criminal fraud case. VII. EVIDENTIARY ISSUES A. Compensation Evidence The defendants challenge the district court’s admission of evidence of their compensation, which the government introduced to prove the defendants’ motive. The compensation evidence included: (1) the defendants’ receipt of company stock when hired; (2) the amount of the defendants’ stock bonuses during the period of the fraud; (3) the defendants’ shares sold during the period of the fraud; (4) the sale price for up to two stock sales per defendant; and (5) other compensation including base salary, cash bonuses, and stock options. We review for abuse of discretion the district court’s evidentiary decisions. United States v. Brown, 415 F.3d 1257, 1264-65 (11th Cir. 2005). “The district court has broad discretion to determine the relevance and admissibility of any given piece of evidence.” United States v. Merrill, 513 F.3d 1293 (11th Cir. 2008). “[E]vidence of wealth or extravagant spending may be admissible when relevant to 28 See Sovereign Military Hospitaller Order of Saint John of Jerusalem of Rhodes and of Malta v. Fla. Priory of the Knights Hospitallers of the Sovereign Order of Saint John of Jerusalem, Knights of Malta, the Ecumenical Order, 702 F.3d 1279, 1291 (11th Cir. 2012) (declining to import Global-Tech’s standard to analyze a fraud claim outside the specific civil patent-infringement context with which Global-Tech was concerned, and stating, “We have been admonished to exercise caution before importing standards from one area of intellectual-property law into another”). 115 Case: 14-12373 Date Filed: 08/11/2016 Page: 116 of 124 issues in the case and where other evidence supports a finding of guilt.” United States v. Bradley, 644 F.3d 1213, 1271 (11th Cir. 2011). A district court has “broad discretion to admit the Government’s ‘wealth evidence’ so long as it aided in proving or disproving a fact in issue.” Id. at 1270, 1272 (finding no reversible error where the district court permitted the government to present substantial evidence of the defendants’ wealth). The district court did not abuse its broad discretion in admitting this evidence. First, the district court carefully limited the wealth evidence to evidence of compensation that depended upon WellCare’s profits. That way, the district court admitted only what was necessary to show that the defendants had an incentive to maximize WellCare’s profits. Second, before the government presented any wealth evidence, the district court instructed the jury to consider such evidence only to the extent it established financial motive for the defendants to commit the charged offenses and for no other reason. The district court further instructed the jury that the defendants’ wealth had nothing to do with whether the defendants were guilty or innocent of the charges against them. The district court’s jury instructions guarded against the chance that the jury would draw any impermissible inferences to the defendants’ detriment. As a result, the district court admitted only relevant evidence and took steps to mitigate any prejudicial effect. 116 Case: 14-12373 Date Filed: 08/11/2016 Page: 117 of 124 Additionally, we reject Kale’s complaint that the government’s wealth evidence was especially prejudicial to his defense because he had only a “modest compensation alongside Mr. Farha’s rich financial rewards.” The record shows that the government presented distinct, individualized evidence of each defendant’s compensation. The admitted evidence showed that Kale’s compensation paled in comparison to Farha’s. If anything, this evidence helped distance Kale from the motive evidence. We find no impermissible spillover effects and no abuse of discretion by the district court. B. Forensic Accountant’s Testimony The defendants argue that the district court abused its discretion under Rule 703 of the Federal Rules of Evidence by allowing Kelly, the forensic accountant, (1) to disclose the fact that WellCare had publicly filed an audited financial restatement with the SEC and (2) to use any of the content of the restatement in his calculations and testimony. Rule 703 addresses an expert witness’s opinion testimony and provides as follows: An expert may base an opinion on facts or data in the case that the expert has been made aware of or personally observed. If experts in the particular field would reasonably rely on those kinds of facts or data in forming an opinion on the subject, they need not be admissible for the opinion to be admitted. But if the facts or data would otherwise be inadmissible, the proponent of the opinion may disclose them to the jury only if their probative value in helping the jury evaluate the opinion substantially outweighs their prejudicial effect. 117 Case: 14-12373 Date Filed: 08/11/2016 Page: 118 of 124 Fed. R. Evid. 703. The rule allows experts to base their opinions on “facts or data” (1) that an expert has “been made aware of or personally observed” or (2) that experts in the particular field would “reasonably rely on.” Those “kinds of facts or data” need not be admissible. In forming his opinions, Kelly reasonably relied on the “facts or data” contained in WellCare’s audited financial restatement. Even though Kelly’s expert opinions themselves were admissible, the defendants challenge his disclosing to the jury (1) the fact that a restatement was filed to correct accounting errors and (2) certain numbers set forth in the restatement. They contend that these facts were inadmissible hearsay. The defendants stress that Rule 703 provides that if the facts or data used by the expert are not admissible, the experts may disclose them only if the probative value “substantially outweighs” the prejudicial effect. Defendants’ argument fails because the financial restatement was admissible as a business record under Rule 803(6). Kelly could both use and reveal this evidence.29 Notably too, the facts and data Kelly disclosed from the financial restatement primarily corroborated his own claims data analyses of WellCare’s over-reporting expenses and under-paying refunds. For context, Kelly’s testimony covered: (1) his own review and analysis of WellCare, Staywell, and HealthEase’s 29 The restatement contained comments and addressed issues beyond the scope of this case. The government offered to redact the restatement, but the district court decided not to admit it. The district court instead allowed Kelly to testify as to how he relied on and used certain financial information in the restatement. If anything, the district court’s careful and practical resolution of this issue underscores how the district court did not abuse its discretion. 118 Case: 14-12373 Date Filed: 08/11/2016 Page: 119 of 124 records, including the claims database; (2) his calculations for each year from CY 2002-2006 as to the differences between the amounts Staywell and HealthEase reported to AHCA and what Staywell and HealthEase actually spent on CMH/TCM services; (3) Kelly’s calculations that the falsely-reported expenses were $29,920,705 more than the actual CMH/TCM expenses; (4) his own calculations of the impact of the falsely-reported expenses on WellCare’s annual financial statements filed with the SEC; and (5) Kelly’s analysis of the inconsistent 80/20 reporting methodologies and Staywell and HealthEase’s use of a results- oriented reporting methodology that started with a predetermined refund amount and worked backward to expense figures to reach that result. After testifying about his own analyses, Kelly explained that public companies like WellCare regularly file 10-K financial statements with the SEC. Sometimes public companies conclude that a filed financial statement contains materially incorrect information. When that happens, the company must file a restatement with the SEC. Kelly told the jury that WellCare had restated its financial statement in 2007 to correct accounting errors related to the refunds required under the AHCA contract. Based on his examination of WellCare’s restatement and Deloitte & Touche’s working papers, Kelly used audited figures from the restatement and calculated that Staywell and HealthEase collectively had owed $35,134,000 more 119 Case: 14-12373 Date Filed: 08/11/2016 Page: 120 of 124 in refunds than what they paid from CY 2002 through CY 2006. Kelly also testified that, for tax years 2004, 2005, and 2006, Staywell and HealthEase’s combined net income before taxes should have been 13.9% lower in 2004, 8.8% lower in 2005, and 6.5% lower in 2006 than they reported. There was no error in admitting Kelly’s testimony about the fact of the audited restatement’s public filing or about certain financial figures in the restatement. The district court correctly noted on several occasions, including when ruling on the defendants’ Rule 703 objection, that the audited restatement qualified as a business record under Rule 803(6) of the Federal Rules of Evidence. The audited restatement was a report made in 2007 in conjunction with a detailed accounting review by those with knowledge of Staywell’s and HealthEase’s books and records. The restatement is a business record of the accounting review itself and its review of what Staywell and HealthEase publicly showed for their eligible expenses during the relevant period of the AHCA contracts. Federal courts commonly admit audited financial reports that restate earnings and are publicly filed with the SEC. 30 See, e.g., SEC v. Jasper, 678 F.3d 1116, 1124 (9th Cir. 2012) (collecting cases). 30 The government argues that any errors as to the restatement should be reviewed for plain error because the defendants either did not properly object or invited the errors by their shifting litigation position. We need not resolve those issues as the district court did not abuse its discretion. 120 Case: 14-12373 Date Filed: 08/11/2016 Page: 121 of 124 Similar to this case, Jasper involved a fraud action in which the SEC alleged that a company’s former CFO perpetuated a fraudulent scheme resulting in the company’s significantly overstating its income. Id. at 1119. The Ninth Circuit rejected the CFO’s argument that the company’s restatement, which the company filed following an internal investigation after the CFO had left the company, could not be admitted under Rule 803(6). Id. at 1122-23. See also In re Worldcom, Inc., 357 B.R. 223, 229 (S.D.N.Y. 2006) (recognizing the admissibility of a financial restatement under Rule 803(6) and stressing the trial judge’s finding that “the intense public scrutiny involved in the restatement of WorldCom’s financial [sic] adequately ensured that the results were trustworthy”). As in Jasper, the restatement is generally admissible under Rule 803(6). The defendants also argue WellCare’s restatement was a result not of the company’s regular practice of a regularly conducted activity but rather of pressure WellCare faced while under threat of criminal prosecution. The Ninth Circuit rejected a similar argument in Jasper, where the CFO argued that the company’s financial restatement should have been excluded because it was “explicitly created with an eye toward pending litigation.” 678 F.3d at 1123. The Ninth Circuit disagreed, stating: This argument has no limiting principle: the filing of an accurate 10-K was and continues to be a legal requirement for Maxim [the company]. In today’s litigation-heavy climate, the filing of any 10-K can always subject companies to legal exposure. That is why lawyers 121 Case: 14-12373 Date Filed: 08/11/2016 Page: 122 of 124 pore over 10-Ks every year at substantial expense to shareholders. Were this court to accept [the CFO’s] contention, virtually every document a public company releases to the public would be inadmissible as a business record merely because companies are worried about litigation risks. That is not the law under the Federal Rules of Evidence. Id., at 1123-24. While the circumstances in which WellCare filed the restatement had legal overtones, the process WellCare used to produce the restatement conformed to regular practice. It is undisputed that Deloitte & Touche conducted an independent audit in accordance with generally accepted accounting principles. 31 As this Court has said, the “touchstone of admissibility under the business records exception to the hearsay rule is reliability, and a trial judge has broad discretion to determine the admissibility of such evidence.” United States v. Langford, 647 F.3d 1309, 1327 31 Deloitte & Touche’s audit report states: We have audited the accompanying consolidated balance sheets of WellCare Health Plans, Inc. and subsidiaries (the “Company”) as of December 31, 2007, 2006, 2005, and 2004, and the related consolidated statements of income, stockholders’ and members’ equity and comprehensive income, and cash flows for each of the four years in the period ended December 31, 2007. . . . We conducted our audits in accordance with the standards of the Public Company Accounting Oversight Board (United States). . . . The defendants argue that the restatement, though publicly filed with the SEC, was not properly authenticated by a custodian and thus could not be used at all. The restatement itself, however, was not admitted and therefore does not require separate authentication. The restatement was discussed only through Kelly’s expert testimony, and the limited content of the restatement that Kelly used was sufficiently reliable and admissible for purposes of his testimony. The defendants were free to cross-examine Kelly on the reliability of the facts and data upon which he relied. 122 Case: 14-12373 Date Filed: 08/11/2016 Page: 123 of 124 (11th Cir. 2011). Any pressure WellCare experienced when cooperating with federal and state law enforcement and government agencies goes to the weight of the restatement’s content rather than to the restatement’s admissibility. Because the financial restatement was audited by an independent accounting firm, was publicly filed with the SEC, and was a reliable and relevant business record, we conclude that Rule 703 does not bar the testimony that Kelly offered regarding the restatement. We also reject the defendants’ allegation that they had no opportunity to alert the jury to circumstances that could cast doubt on the restatement’s reliability. 32 Defense counsel asked Kelly if WellCare had filed its restatement “after the company had been raided by 200 agents.” Kelly responded, “Yes, it was.” Defense counsel continued, “It was when the company was under federal investigation; right?” Kelly responded, “Yes, that’s true.” Defense counsel continued to press, “Well, the restatement was done at a time when the company was under threat of criminal prosecution; correct?” Kelly again, “I believe that’s right.” Later defense counsel asked, “[S]ometimes companies restate in order to survive; isn’t that correct?” Kelly answered, “Sometimes companies restate, yes, to correct their financial statements.” He explained that be it “for purposes of 32 Several months before trial, the district court expressed a preliminary willingness to allow the restatement to be admitted into evidence. After defense counsel objected, the court said, “Well, you can notify your forensic accountant that’s maybe an area where the government’s headed.” The government’s expert forensic accountant’s use of the audited financial statement was therefore no surprise. 123 Case: 14-12373 Date Filed: 08/11/2016 Page: 124 of 124 getting lending or whatever . . . accurate financial statements are very important.” Kelly finished, “So, yes, companies do that to survive sometimes.” The defendants successfully communicated to the jury that external legal pressure may have motivated WellCare to restate its financials. By the time Kelly discussed the restatement and certain numbers therein, the jury already had heard about the false expense reports from Kelly, West, and internal company records. Kelly’s additional calculations based on the restatement’s financial information, while probative, were cumulative. The district court did not abuse its discretion in allowing Kelly to testify to a limited extent about the restatement.33 VIII. CONCLUSION For all of the forgoing reasons, we affirm the defendants’ convictions. AFFIRMED. 33 Kale alone contends that the district court improperly limited his impeachment of Pearl Blackburn’s testimony and improperly allowed the government’s hypothetical question to witness Michael Turrell. Kale has not shown any reversible error as to these issues. 124