Opinion
19 CR 864
03-21-2024
United States of America, v. Rishi Shah, Shradha Agarwal, Brad Purdy, and Ashik Desai.
MEMORANDUM OPINION AND ORDER
Honorable Thomas M. Durkin, United States District Judge.
Rishi Shah, Shradha Agarwal, and Brad Purdy (“Defendants”) have filed motions for judgment of acquittal on all counts for which they were found guilty pursuant to Federal Rule of Criminal Procedure 29, and motions for a new trial pursuant to Federal Rule of Criminal Procedure 33. R. 486; R. 487; R. 489. They challenge the sufficiency of the evidence against them. Because the evidence is sufficient to support the jury's verdict, and none of Defendants' arguments for a new trial are meritorious, the motions are denied.
Background
Defendants are former executives at Outcome Health (“Outcome”), a company that placed television screens, tablets, and wallboards that displayed educational content in doctor's offices and sold advertising space on those devices to pharmaceutical companies and other clients. The superseding indictment charged that from 2011 to at least 2017, Defendants and another Outcome employee, Ashik Desai, knowingly and intentionally devised a scheme to defraud Outcome's clients, lenders, and investors. R. 14. Specifically, the indictment alleged that Defendants falsely represented to Outcome's clients that the company's network included specific doctors and doctor's offices that the clients were targeting for advertising; lied to clients about how many screens the clients' advertisements were running on; falsely inflated patient engagement metrics associated with Outcome's tablets; caused material under-delivery on advertising campaigns; caused clients to pay for advertising that was not delivered; caused the material inflation of revenue in Outcome's financial statements; and used Outcome's inflated financial statements to obtain a $110 million loan in April 2016, a $375 million loan in December 2016, and $487.5 million in equity financing in 2017. Id. The indictment further alleged that to conceal that scheme, Defendants marginalized, undermined, and ignored whistleblowers who raised concerns; at times siloed information by directing Outcome employees not to include salespeople who were interacting directly with the clients on certain communications; discouraged clients from conducting their own campaign performance studies and instead encouraged the use of Outcome's chosen measurement company; and hid under-delivery on advertising campaigns from the auditor, investors, and lenders. Id.
Desai pled guilty on December 9, 2019. R. 45; R. 48.
Defendants were charged with mail fraud, wire fraud, and bank fraud, Shah was also charged with money laundering, and Purdy was also charged with making a false statement to a financial institution. Id. Defendants pled not guilty to all counts, and proceeded to an eleven-week-long jury trial. Over the course of nearly nine weeks of evidence presentation, the jury heard from former Outcome employees Desai, Sameer Kazi, Jason Ketchum, David Ma, Collin Williams, and Bridget O'Donnell; pharmaceutical company victims and advertising agency representatives Yesenia Bautista (Target Health), Matthew Ubriaco (Zenith), Tameka Teal (Compas), Courtney Cruz (Pfizer), Joseph DiFoglio (Carat), Lyndon Chin (Assembly), Tom Coyle (Initiative), Lynn Meier (Starcom), and David Dobbins (Boehringer Ingelheim); lender and investor victims Ken Eberts (Goldman Sachs), James McHugh (JPMorgan Chase), Todd Cozzens (Leerink), and Laela Sturdy (CapitalG); auditor Tracy Harrison (Deloitte & Touche LLP (“Deloitte”)); expert Michael Petron; and FBI Special Agent Megan Poelking. Shah and Purdy called one witness: data analyst Amanda Malusky Krauss. Agarwal called no witnesses.
At the close of the government's case and again at the close of the defense case, Defendants moved for a Rule 29 judgment of acquittal, and those motions were taken under advisement. On April 11, 2023, after nearly three days of deliberations, the jury returned a verdict of guilty for Shah on Counts 1, 2, 4, 5, 9-19, 22, and 24-26, for Agarwal on Counts 1, 2, 4, 9, 11, 13-18, 22, and 24-26, and for Purdy on Counts 1-4, 7-9, 13, and 22-26. R. 447. The jury returned a verdict of not guilty for Shah on Counts 20, 21, 23, for Agarwal on Counts 20 and 23, and for Purdy on Counts 11 and 21. Id. Defendants each renewed their motions for acquittal pursuant to Rule 29 or, in the alternative, for a new trial pursuant to Rule 33. R. 486; R. 487; R. 489. This ruling applies to both the pre- and post-verdict Rule 29 motions.
Discussion
I. Motions for Judgment of Acquittal
Rule 29 provides that “the court on the defendant's motion must enter a judgment of acquittal on any offense for which the evidence is insufficient to sustain a conviction.” Fed. R. Crim. P. 29(a). In reviewing a challenge to the sufficiency of the evidence, courts must “afford great deference to jury verdicts, view the evidence in the light most favorable to the jury's verdict, and draw all reasonable inferences in the government's favor.” United States v. Brown, 973 F.3d 667, 681 (7th Cir. 2020). Courts “do not re-weigh the evidence or second-guess the jury's credibility determinations.” United States v. Taylor, 637 F.3d 812, 815 (7th Cir. 2011). Put differently, a court will only overturn a jury's verdict “if the record contains no evidence, regardless of how it is weighed, from which the jury could find guilt beyond a reasonable doubt.” United States v. Fitzpatrick, 32 F.4th 644, 650 (7th Cir. 2022). That burden is “nearly insurmountable.” Id. at 649. Additionally, a challenge to the sufficiency of the evidence proving intent “is exceedingly difficult to win.” United States v. Dingle, 862 F.3d 607, 614 (7th Cir. 2017).
A. Mail Fraud, Wire Fraud, and Bank Fraud Counts
Defendants were each convicted of several counts of mail fraud (18 U.S.C. § 1341), wire fraud (18 U.S.C. § 1343), and bank fraud (18 U.S.C. § 1344). In order to prove Defendants guilty of mail or wire fraud, the government had to prove the following elements beyond a reasonable doubt: (1) that the defendant knowingly devised or participated in a scheme to defraud; (2) with the intent to defraud; (3) that the scheme involved a materially false or fraudulent pretense, representation, or promise; and (4) that the defendant used or caused the use of the United States Mails or caused interstate wire communications to take place for the purpose of carrying out the scheme or attempting to do so. R. 444 at 23 (Instruction No. 22); Seventh Circuit Pattern Criminal Jury Instructions at 538. Further, in order to prove Defendants guilty of bank fraud, the government had to prove the following elements beyond a reasonable doubt: (1) that there was a scheme to defraud a bank; (2) that the defendant knowingly executed or attempted to execute the scheme; (3) that the defendant acted with an intent to defraud; (4) that the scheme involved a materially false or fraudulent pretense, representation, or promise; and (5) that at the time of the offense, the Federal Deposit Insurance Corporation (“FDIC”) insured the deposits of the bank. R. 444 at 235 (Instruction No. 23); Seventh Circuit Pattern Criminal Jury Instructions at 635.
Relevant to all three charges, intent to defraud requires proof that the defendant “act[ed] knowingly with the intent to deceive or cheat the victim in order to cause a gain of money or property to the defendant or another, or the potential loss of money or property to another.” R. 444 at 30 (Instruction No. 28); see also United States v. Faruki, 803 F.3d 847, 853 (7th Cir. 2015). Additionally, a scheme is “not limited to an isolated instance of conduct” but rather understood as a “continuing course of conduct, during a discrete period of time.” United States v. Thomas, 986 F.3d 723, 729 (7th Cir. 2021). Moreover, since the elements for wire fraud directly parallel those for mail fraud, the Seventh Circuit treats “‘cases construing one [as] equally applicable to the other.'” United States v. Kelerchian, 937 F.3d 895, 909 n.2 (7th Cir. 2019) (quoting United States v. Leahy, 464 F.3d 773, 786 (7th Cir. 2006)).
The scheme charged in Count 1 was incorporated into each of the other counts.
1. Defrauding Pharmaceutical Companies (Counts 1, 2, 4, 5, 11, 23, and 25)
Defendants argue there is insufficient evidence of their knowing participation in the scheme to defraud or intent to defraud the pharmaceutical companies or their advertising agencies (collectively, “pharma-clients”). The relevant counts are as follows:
- Count 1 charged Defendants with the February 2, 2015 mailing of a $53,790 check from Pfizer (GX 438) to pay Outcome's November 1, 2014 invoice for the Xeljanz campaign (GX 410).
- Count 2 charged Defendants with the February 4, 2015 mailing of a $229,170 check from AbbVie (GX 439) to pay Outcome's November 1, 2014 invoice for the Humira campaign (GX 409).
- Count 4 charged Defendants with the May 14, 2015 mailing of a $89,100 check from Biogen (GX 466) to pay Outcome's February 1, 2015 invoice for the Tecfidera campaign (GX 437).
- Count 5 charged Shah with the November 14, 2015 email from Shah to Dave DiCosola and Desai stating in part, “[Y]ou have to make sure not to talk about total installs in the everyone email - this is a big deal because sponsorship is on there too and they're selling much larger quantities for next year.” GX 494.
- Count 11 charged Defendants with the July 22, 2016 mailing of a $1,071,202.56 check from Boehringer Ingelheim (GX 597) to pay Outcome's June 1, 2016 invoice for the Jardiance campaign (GX 582).
- Count 23 charged Defendants with the March 16, 2017 mailing of a $458,261.62 check from Merck (GX 897) to pay Outcome's December 1, 2016 invoice for the Belsomera campaign (GX 675).
- Count 25 charged Defendants with the April 24, 2017 mailing of a $566,799 check from Pfizer (GX 930) to pay Outcome's February 1, 2017 invoice for the Xeljanz campaign (GX 836).
Only Shah and Agarwal were found guilty of this charge, and thus they are the only defendants who challenge the sufficiency of the evidence on this count.
Only Purdy was found guilty of this charge, and thus he is the only defendant who challenges the sufficiency of the evidence on this count.
A review of the evidence illustrates how Defendants knowingly and intentionally engaged in a scheme to defraud Outcome's pharma-clients through overselling Outcome's inventory and concealing pervasive under-deliveries. At the start, by nature of its business, Outcome faced a “two-sided market.” GX 70. That is, Outcome needed to install enough screens in doctor's offices to attract advertisers, while also generating enough advertising revenue to fund the installation of new screens. That aspect of Outcome's business precipitated the first part of the scheme: inflating list matches. Pharma-clients sent Outcome lists of doctor's offices they wanted to target for advertising campaigns. Outcome compared the lists with its inventory of offices where screens were installed and generated a “list match.” Trial Transcript (“Tr.”) 2167. But instead of using current inventory, Outcome often used “projected inventory” to entice clients to buy larger, more expensive advertising contracts. Tr. 2170, 2172.
The evidence showed that in the early days of the company, Shah and Agarwal were heavily involved in day-to-day sales and promoted the use of projected inventory. For example, with Shah's approval, a salesperson emailed a client in February 2011 that Outcome was “currently in the waiting rooms of 1/3 of all endocrinologists in the US.” GX 2. When Agarwal asked whether that assertion was “inaccurate,” Shah acknowledged that as a standalone statement, it was. Id. And when Agarwal pointed out that Shah had made that representation “a lot,” Shah added, “Let's just get to [the claimed number] quickly.” Id. In November 2012, Agarwal herself, with Shah's awareness, directed an employee to conduct a list match using projected inventory, GX 67, and one year later, admitted that the list the client had wasn't the “true list” as Outcome did not have even one-third of the inventory “installed at the time,” GX 265.
The evidence also demonstrated that by 2013, Purdy became involved, albeit to a lesser extent, in the effort to sell pharma-clients on “the future,” i.e., projected inventory. Outcome employee David Ma testified that Purdy taught him how to conduct list matches using methods that would sweep in offices where Outcome did not have installed screens. Tr. 2184-87. Ma also recounted that when he expressed his concern that these methods “would yield inaccurate results” and “overrepresent [Outcome's] inventory,” Purdy shrugged, “that's just kind of the way we do it,” and “I think it's okay.” Id. 2187-88.
To be sure, some pharma-clients were content with buying projected inventory, either through growth schedules or weighted average delivery written into the contracts. But the evidence showed that growth and weighted average contracts were the exception, not the norm. E.g., id. 2202, 2398, 5492. And testimony from pharma-clients revealed that they intended to buy existing inventory and were generally not told that the list matches did not reflect Outcome's current inventory. Id. 2031-32, 6094, 6256, 8105.
Generally speaking, “weighted average” contracts promised delivery of a weighted average number of offices or devices over the duration of the contract rather than a specific number of offices or devices per month. For these contracts, Outcome would bill clients for the same amount each month, even if the actual delivery for that month exceeded or fell short of the weighted average amount. “Growth” contracts promised delivery for a certain number of offices or devices at the start of performance and increasing numbers over the duration of the contract. For these contracts, Outcome would bill clients for the specific growth targets set by the contract. See Tr. 609-10, 864-65, 2397-98.
The evidence showed that Defendants helped conceal that fact by siloing information about how the list matches were created away from the salespeople who directly interfaced with pharma-clients. For example, Shah wrote to Agarwal directly asking whether a salesperson knew a certain list match was projected, removing that salesperson from the email. GX 140. Agarwal responded that while she wasn't sure in that specific case, “generally, the team doesn't know it's a projection because they get confused about how to represent it, even though they need not say it any differently.” Id. The evidence featured additional conversations between Defendants of a similar nature. See GX 494 (Count 5) (Shah instructing Dave DiCosola “you have to make sure not to talk about the total installs in the everyone email - this is a big deal because sponsorship is on there too and they're selling much larger quantities for next year”), 274 (Agarwal telling Desai to take salespeople off email chains when discussing “what data to use,” noting “I've noticed their confidence level in our data change dramatically when presenting to clients if they believe it's accurate vs made up.”), 1099 (Purdy stating “we should probably make some sort of barrier between the [sales] team and ops with regards to tablets and/or take off the cumulative counts. With all the projecting I want [sales] to stay focused on selling the future and realized that the below [data] may even be inaccurate for them[.]”); see also Tr. 2175, 341415, 3429-30, 3436, 3473-74, 5501.
Other evidence indicated that on numerous occasions, Defendants knew that pharma-clients were being misled. For instance, in August 2013, the same month that Desai began working at Outcome, Shah, Agarwal, and Desai discussed an email from Boehringer Ingelheim about “ensuring proper ROI [return on investment] measurement” of the Pradaxa campaign. GX 263. After Agarwal admitted that the Pradaxa list match wasn't the “true list” as Outcome did not have even one-third of the inventory “installed at the time” the list was provided, Shah added that if the client started asking questions about the list, they could shift to a weighted average approach. Id. Later in the month, Desai told Ketchum that he had spoken to Shah about the Pradaxa list and shared, “In the end, we're open to being honest if it's caught[.]” GX 284. Along the same lines, in October 2013, Shah, Purdy, and Desai discussed presenting a list of offices to Novo Nordisk that included a number of offices not installed at the time as if they were current inventory. GX 322, 324; Tr. 3421-22, 3596-3603.
Some of Outcome's contracts included ROI guarantees, which promised a certain amount of return for every dollar spent on advertising. See Tr. 2400-01 (“If a [pharma-client] is paying us $1 million and they require 3:1, that means that we had to deliver $3 million in value.”). Outcome generally contracted with third party vendors to conduct studies of ROI, such as IMS. See id. 2414-19.
Ma testified that in early 2015, he shared his concerns about selling inventory Outcome did not have and Agarwal's “jarring” response were words to the effect of, “we throw smoke bombs . . . to clean up our operations.” Tr. 2422-25, 3030; see also GX 265. Agarwal's comments echoed Shah's remarks to a room full of Chicago entrepreneurs in November 2012 that Outcome “threw a smoke bomb”: Outcome had doctor's offices “commit to taking the installation” of a device, while having advertisers “commit to buying the network,” acknowledging “You cannot be late. Otherwise it's fraud. Right? I mean it . . . You know, you're selling something you don't have.” GX 70.
Because Outcome was “selling devices that were not there,” it “had significant under-delivery on a number of [its] programs, even while [it was] billing [its] clients as if [it was] delivering in full. Tr. 3431; see also Tr. 2200. The evidence demonstrated that under-deliveries-coined “deltas”-were pervasive, recurring, and well-known by Defendants. There was evidence that Shah, Agarwal, and Purdy were each aware of ongoing under-deliveries, first through emails and small groups and then through formal “delta reports” and “growth models” shared with Outcome's senior leadership and discussed in off-site retreats. See, e.g., GX 357, 360, 397, 441, 442, 460, 461, 624, 1027, 1098, 1153; Tr. 2211-12, 2237-39, 2254, 2261-62, 2273, 2275, 3653-57, 376869. In March 2016, Shah acknowledged the recurring nature of deltas when he noted, in an email to Agarwal and his chief of staff, that he didn't “want to spend every meeting” with Outcome's executive team [including Desai and Purdy] on the issue of deltas. GX 557. He added that he didn't want to hear “new ideas not cleared by me,” concerned that it would turn into an “inquisition.” Id.
Yet, testimony from the pharma-clients revealed that they were not made aware of under-deliveries. See, e.g., Tr. 2032, 2200-01, 3313, 3519, 6094. Some evidence indicated that pharma-clients were unaware because of active concealment at Defendants' direction or with their approval. The jury saw several examples of Desai working with Defendants to conceal the deltas from pharma-clients by reassigning inventory from one client to another, e.g., GX 330, 337, 1928; inflating patient-engagement data on Outcome's tablets, e.g., GX 345, 432, 1144, 1908; Tr. 3024; and filtering data used in ROI studies, e.g., Tr. 3568-69, 5514-15. Later, when Outcome was faced with even larger deltas, Desai began changing performance data received from the third-party vendor IMS such as “the total number of doctors studied, the percentage lift, the ROI” before sharing it with clients. Tr. 3870-71, 3883-84, 3917. By November 2016 at the latest, Shah and Agarwal were aware of reports by a salesperson named David Jundt that Desai was changing ROI data. GX 673, 747; Tr. 3916-17. Indeed, they discussed with Desai how to address ongoing talks between the offices about the legitimacy of the ROI data. GX 673. But neither Shah nor Agarwal ever confronted Desai about the accusation. Tr. 3925.
All the while, Outcome continued billing its clients as if it had delivered what was asked. Tr. 2210-11, 3431, 3651-52. The evidence, viewed in the light most favorable to the verdict, further indicates that Defendants knew this was happening. Desai testified that he discussed revenue at meetings with Shah and Agarwal and sent all three Defendants regular revenue reports, showing Outcome meeting its revenue targets and some including campaign-specific billing data. E.g., Tr. 3652, 3831-32, 3833, 3838-40; GX 1035, 1036, 1037, 1038, 1039, 1040, 1041, 1043, 1044, 1044a, 1045, 1046, 1047, 1048. Further, despite frequent reminders about massive inventory shortages in certain therapeutic categories, like rheumatology and neurology, Defendants knew that Outcome continued to sell larger advertising campaigns, with growing deltas, year after year. GX 107, 136, 193, 205, 221, 330, 357, 359, 461, 461a, 594, 594a, 704, 1098, 1153. Moreover, Shah supplied the revenue goals, Tr. 3844, and pushed for increasing sales goals despite deltas, id. 3027-29. See also id. 2348-49, 3033, 3628-29.
In June 2016, Shah, Agarwal, and Desai encountered a “massive” issue, where there was “essentially no delivery” of advertising for two months of a campaign for the brand Invokana. GX 588, 589 (Shah texting Agarwal that “all of [the campaigns] may be totally under delivering”). A few weeks later, the three received the results of an internal audit of what Outcome was “delivering versus what was in the contract” for “2016 programs.” Tr. 3781-82; GX 594. The audit revealed under-deliveries of 10% or more on 77 of 137 active campaigns and six campaigns with “dark” periods during which no advertisements were played, even though clients were billed in full. GX 594, 594a; Tr. 3782. The audit further found 31 exam-room tablet and 12 waiting-room TV campaigns with under-deliveries of 50% or more. GX 594a. Yet, none of the underdeliveries were disclosed to the pharma-clients. Tr. 3787-88.
Several months later, on Christmas Eve 2016, Shah sent Purdy and Desai a Voxer (a voice memo sent between employees) summarizing “grave concerns” that Purdy had shared about Outcome's “ability to fulfill the current standing of 2017 contracts.” GX 701. Desai replied two days later, observing how Outcome had, for years, sent monthly proof-of-performance affidavits to clients, which falsely attested all contractual requirements had been met despite recurring under-deliveries. GX 704. Desai recommended they “change this practice,” so that affidavits are a “source of truth as to what, in fact, ran in the 30-day period.” Id. He also noted that, for “a longtime [therapeutic] category” like rheumatology where Outcome had been underdelivering on campaigns for years, “starting to call attention to massive inventory gaps” by suddenly submitting accurate affidavits would “open up a whole and historical legacy of potential issues,” including recalling revenue and “more drastic impact[s]” like losing clients entirely. Id. In response, Shah tasked Purdy and Desai with finding a solution that balanced “the goal[s] . . . of correctly reporting on affidavits while not . . . causing a bunch of account-level issues.” GX 707.
Viewing this evidence altogether and drawing all reasonable inferences in the government's favor, a jury could conclude that Defendants knowingly participated in a scheme to defraud and intended to defraud Outcome's pharma-clients. Nonetheless, Defendants contend that the convictions on these counts cannot stand.
First, they claim that a rational jury could not have found them guilty of these charges without evidence they were specifically aware of (1) the terms of each contract in question and whether it allowed for growth or weighted average delivery; (2) what Outcome delivered on each contract during the relevant time period; (3) that Outcome was providing each client with false proof of performance documents; and (4) what Outcome billed each client during the relevant time period. See R. 489 at 5; R. 486 at 6-9, 11. But those requirements are found nowhere in the case law cited by Defendants. Instead, the jury had to find beyond a reasonable doubt that Defendants knowingly participated in the scheme with the intent to defraud. And here, as described, there was significant evidence that Defendants knew and intended to deceive or cheat pharma-clients. Considered in the light most favorable to the verdict, the evidence showed that Defendants knew that Outcome was consistently inflating list matches, under-delivering on its contracts, and concealing that under-delivery through a variety of means. That there may not have been evidence that Defendants knew the precise terms of each contract and delivery and billing data does not mean that they did not knowingly participate in the scheme with the intent to defraud Outcome's clients. Further, the jury was properly instructed that they had to find beyond a reasonable doubt that each Defendant acted with the intent to defraud. R. 444 at 23, 30, 42.
Relatedly, Defendants assert that the government invited the jury to speculate about their knowledge and intent. Defendants claim there was no evidence that they knew that clients were billed in full despite under-deliveries and knew there had been no make goods in 2015. It is true that the Court “must take special care to guard against the possibility that a defendant may be found guilty by either speculation or mere association.” United States v. Garcia, 919 F.3d 489, 503 (7th Cir. 2019). But that's not what happened here. Instead, the jury made a reasonable inference based on Defendants' awareness of ongoing and pervasive under-deliveries and their simultaneous knowledge of Outcome's increasing revenue quarter after quarter through Desai's revenue reporting. If Outcome was only billing on what was delivered, it would not have met its revenue targets. Likewise, Desai testified that there were no make goods in 2015 and when Outcome began giving make goods in late 2016, Shah and Agarwal talked about them extensively. The reasonable inference from the absence of any conversations between Defendants about make goods in 2015, in stark contrast to 2016, is that they knew there were no make goods in 2015.
“Make goods” refer to offers made to clients to compensate, or “make good,” for what was under-delivered. It may take the form of a credit back to the client or running advertising for a longer period of time or at additional offices at no cost. See, e.g., Tr. 2036, 2941.
Defendants recast the evidence in a number of ways to argue that there was insufficient proof of their knowing participation in the scheme and intent to defraud. Primarily, they maintain-as they did at trial-that the fraudulent scheme was wholly of Desai's making and hidden from them. Similarly, Shah claims that the evidence at most shows that he had a general sense that Outcome occasionally had “operational errors.” GX 489 at 5. But contentions “that the evidence supports equally rational inferences are unavailing, giving this court's obligation to view the evidence in the light most favorable to the government.” United States v. Moede, 48 F.3d 238, 241 (7th Cir. 1995); see also United States v. Huels, 31 F.3d 476, 479 (7th Cir. 1994). In addition, this argument runs headlong into the evidence showing Defendants' involvement in manipulating list matches before Desai arrived at the company. E.g., GX 67, 265. While there was evidence that Desai kept the fact that he changed ROI data from Defendants, there was other evidence showing how Desai directly collaborated with Defendants and sought their guidance in executing the scheme.
Desai testified at length about how he followed Defendants' model and conducted the fraud with their direction and approval. That testimony did not stand alone. It was extensively corroborated by other evidence, including the testimony of Ma regarding Defendants' knowledge and lies to clients (e.g., Tr. 2371, 2393, 3036, 3396, 3927); the testimony of Ma and pharma-clients undermining the inference that Defendants believed Outcome routinely entered growth or weighted average contracts (id. 2202, 2398, 3173-75, 3279, 6043-44, 6049, 6092, 6144, 6241-42); communications showing the general practice before Desai's arrival of siloing information from salespeople (e.g., GX 140, 183, 274, 326, 494); Desai's grand jury testimony (Tr. 5649-73); and the emails, Voxers, and text messages shown to the jury that matched Desai's assertions. A large part of the defense case was to portray Desai as a master criminal and that he was lying about Defendants' involvement in the fraud. The jury apparently found Desai credible, and it is not the province of this Court, in weighing a Rule 29 motion, to “reweigh the evidence or ‘second guess'” that determination. United States v. LeBeau, 949 F.3d 334, 346 (7th Cir. 2020) (quoting United States v. Coscia, 866 F.3d 782, 795 (7th Cir. 2017)).
Additionally, Defendants highlight how the government largely ignores Ketchum's testimony. But the favorable testimony they highlight reveals the pitfall of their argument. That Ketchum answered “yes” or “correct” to several helpful propositions on cross-examination after reviewing select documents does not do the work that Defendants claim. E.g., Tr. 1342 (answering “correct” that certain emails shown on direct examination did not include any instruction from Shah to tell a client that the list match had projections), 1325 (answering “yes” to whether he would agree that in “2012, back when Mr. Shah was in the room with the clients, Outcome is explicitly disclosing how many sites it is currently in and what it projects to be in the future”). Rather, Ketchum's willingness to answer “yes” dozens of times and to nearly every question posed by both the government and defense suggests that the jury in all likelihood did not find him credible. As previously stated, it is not the job of this Court to “second guess” that conclusion. LeBeau, 949 F.3d at 346.
In its simplest form, Defendants engaged in a scheme to over-sell, underdeliver, and conceal under-delivery from Outcome's pharma-clients. That is fraud when done knowingly with the intent to deceive or cheat the victims, and there was ample evidence that Defendants acted with that intent. The defense in this case was not that there was no fraud. It was that it was Desai's fraud that he executed with his team and concealed from Defendants. Desai and Ma said otherwise. They were thoroughly cross-examined, and their motives were extensively picked apart. The jury apparently found them credible, and that conclusion is not surprising given the amount of evidence corroborating their testimony. All told, the evidence is sufficient to sustain the convictions on the counts related to defrauding pharma-clients.
2. Wire Fraud as to the May 4, 2015 Email (Count 3)
Purdy also challenges the sufficiency of the evidence on Count 3, which charged him with wire fraud in connection with an email exchange with Ma and Desai on May 4, 2015. Around that time, Purdy was preparing to pitch a hospital system about installing Outcome devices at its facilities. Tr. 2383. For the pitch, he asked Ma to send him tablet engagement metrics that were reported to pharma-clients. Ma cautioned Purdy that the metrics were “not real” and asked whether Purdy was “comfortable sharing non-real stats,” to which Purdy confirmed he was. DX 3456. When Ma emailed the data to Purdy and Desai, he again flagged that the metrics were “inflated.” GX 465. Purdy followed up asking for the false data in a presentation deck form, DX 3456, and Ma sent a deck that was created for a presentation to one of Outcome's pharma-clients, GX 465. The email transmitting the deck, including Ma's previous warning that the data was “inflated,” is charged as Count 3.
Purdy argues that this email was not in furtherance of the scheme because he was collecting the data to pitch health systems that might join Outcome's network, not pharma-clients. But the charged wire is not an email from Purdy to the health systems. Instead, the email is an internal discussion between Purdy and Ma about how pharma-clients were receiving fake, “inflated” data. Indeed, the deck that Ma sent was created for a presentation to a specific pharma-client. To be “in furtherance” of the fraud, it is enough that an email is “‘incident to an essential part of the scheme.'” See United States v. Sheneman, 682 F.3d 623, 629 (7th Cir. 2012) (quoting Schmuck v. United States, 489 U.S. 705, 710 (1989)). Viewed in the light most favorable to the verdict, this internal discussion about how pharma-clients were being deceived through inflated tablet metrics was, at the very least, “incident to” concealing under-delivery from pharma-clients. Moreover, the fact that Purdy responded without shock or surprise to Ma's multiple comments that the data being given to pharma-clients was false further supports Purdy's knowing participation in the scheme and his intent to defraud pharma-clients. For those reasons, the evidence is sufficient to sustain the conviction on Count 3.
3. Silencing Whistleblowers (Counts 14-19)
Shah and Agarwal argue that there is insufficient evidence that they silenced whistleblowers as part of the scheme. The relevant counts are as follows:
- Count 14 charged Shah and Agarwal with the January 23, 2017 Voxer from Agarwal to Shah, which provided in part: “[Sameer Kazi] said that we are operating an unethical business . . . that [Desai] is changing numbers . . . It sounds like he may even share this with others . . . if [Adam Prowker and Lianne Pierce], in [Kazi's] first week, are coming to him and sharing this, that is not very good leadership on their part and we may just have to clean up in a lot of different areas of that org.” GX 747.
- Count 15 charged Shah and Agarwal with the January 24, 2017 iMessage from Shah to Agarwal and Desai, which provided: “My sense is you potentially quickly cycle [Kazi] and [Prowker] and [Pierce] out and bring in people who can fix the issues.” GX 778.
- Count 16 charged Shah and Agarwal with the January 24, 2017 Voxer from Shah to Agarwal, which provided in part: “Did [Madan Nagaldinne] share how the conversation ended . . . It seems striking that he would say things like, ‘once this goes public,' or ‘Once I tell [headhunter], once I tell [another Outcome executive].' And I'm a little nervous about not firing him tomorrow if I can't figure out that he's really trustworthy.” GX 771.
- Count 17 charged Shah and Agarwal with the January 24, 2017 Voxer from Nagaldinne that Agarwal forwarded to Shah, which provided in part: “he opened up the spreadsheet and he said, ‘Take a look at this,' and he shows me all' of this, that he has kind of done this analysis with a bunch of people . . . and then there are at least 25, 30 probably even more contracts where we are not hitting those ROI guarantees of 3 to l or 2 to l . . . then he brought up the same issue of the affidavit again . . . saying that that is fraud . . . He kind of said, ‘This is a humongous operational problem that we have. This is terrible.' . . . ‘Those contracts where you have collected something and we haven't been able to deliver otherwise.' And then he said, ‘I am concerned that more than 50% of our revenue can be recalled....” GX 760.
- Count 18 charged Shah and Agarwal with the January 24, 2017 iMessage from Shah to Agarwal, which provided, “It sounds like [Kazi] has been walking around all day affronting people on affidavits et al[.]” GX 778.
- Count 19 charged Shah with his February 1, 2017 email to Desai, forwarding a letter from an attorney representing Prowker, which provided in part, “Based on information available to [Prowker], including the data reports from [the measurement company], it appears that the Company is artificially inflating the performance results it reports to clients to create the false appearance that it is providing content to more physician offices than is actually the case and is also inflating actuals by measuring offices that are not live[.]” GX 1911.
A review of the events surrounding the charged communications illustrates how they were made for the purpose of executing the scheme, and further supports Defendants' knowing participation in the scheme.
Sameer Kazi began working at Outcome on January 9, 2017 as the COO for Outcome's life sciences group, and quit two weeks and three days later. GX 1123; Tr. 215-25, 231-32, 237. During his second week, he met with two employees he supervised, Adam Prowker and Liane Pierce, who told him that Outcome sold inventory it did not have, sent clients monthly affidavits with falsely inflated inventory and delivery figures, and manipulated ROI results before sending them to clients. Tr. 277-80. Concerned about revenue being called back and that “this was at least to some degree fraudulent,” id. 282, Kazi promptly scheduled a meeting with Shah for January 25, 2017. He told Outcome's head of human resources, Madan Nagaldinne, about the issues he planned to raise, who in turn shared those concerns with Agarwal. GX 747; Tr. 308-09.
Agarwal immediately reached out to Shah in a Voxer, relaying Kazi's claim that they were “operating an unethical business” and acknowledging that it “wasn't the first time” they had heard about Desai “changing numbers.” GX 747 (Count 14). She noted that “it sounds like [Kazi] may even share this with others,” and flagged that if Prowker and Pierce were raising this issue in Kazi's first week, it wasn't “very good leadership on their part” and they might need to “clean up” that area of the company. Id. Later that day, Shah sent a Voxer to Agarwal expressing his concern that Kazi might “go[] public” with his concerns, adding, “I'm a little nervous about not firing [Kazi] tomorrow if I can't figure out that he's really trustworthy.” GX 771 (Count 16).
The next day, Shah checked back in with Agarwal for any updates from Nagaldinne. GX 779. Agarwal forwarded him Nagaldinne's Voxer, which relayed Kazi's concerns that “it is fraud” and they risked more than 50% of the revenue being recalled because of ROI misses. GX 760 (Count 17). Shah sent that Voxer to Desai, GX 760a, and later texted Agarwal and Desai about how Kazi had “been walking around all day affronting people with affidavits et al,” GX 778 (Count 18). Shah suggested that they “quick cycle [Kazi, Prowker, and Pierce] out and bring in people who can fix these issues,” id. (Count 15); see also GX 780A (Agarwal messaging Shah on January 25, 2017 that Prowker was refusing to turn decks “because all the numbers are [‘]misrepresentations[']” and Shah responding that he thought Prowker had “very strong philosophical disagreements” and needed “to be fired immediately”).
Shah met with Kazi on January 25, 2017. Shortly after the start of the meeting, Shah claimed there was a “cultural misfit” between Kazi and Outcome and that Shah “had received several reports of people stating that [Kazi] was a ‘bull in a China shop' poking his nose into parts of the business.” Tr. 309-10. Kazi tried to redirect the conversation toward his concerns, namely “inventory shortfall,” “misrepresentation of inventory on affidavits,” and “misrepresenting [ROI] results and inflating them.” Id. 310-11. Shah did not respond with alarm, instead noting that he was aware of these issues, and showed no interest in discussing them with Kazi. Id. 312-13, 315. Based on this response, Kazi decided to quit immediately. Id. 315. Shah sent a Voxer to Purdy, describing how Kazi was not a “cultural fit,” recounting Kazi's concerns about affidavits, ROI, and forecast, and commenting, “if you talk to people who were at Google or Facebook from, you know a hundred million to a billion [in revenue], they will tell you these are very normal things . . . Google data was false and totally crazy in the foreign markets in [the] early years.” GX 785.
Days after Kazi's departure, Prowker requested a meeting with Shah to discuss his “concerns about how the company was doing business.” Tr. 4090. At the meeting, Prowker described feeling “on the defensive as of late” with Kazi's quick exit and his belief that some of the “contributing factors” for Kazi's departure were conversations with Prowker and Pierce about company operations. GX 818; Tr. 4092. In response, Shah called Prowker “dangerous,” and the conversation devolved into an argument. GX 818. Two days later, Shah told Desai that Prowker was “basically threatening to sue us and make it public with clients” and that he was “just gonna fire this guy.” GX 831. The following afternoon, Prowker's lawyer emailed Shah and Outcome's general counsel, Collin Williams, a whistleblower letter, GX 1911 (Count 19), which was forwarded to Purdy. That letter accused Outcome of “fraudulent” business practices, including: (1) “artificially inflating the [ROI] results it reports to clients to create the false appearance that it is providing content to more physician offices than is actually the case”; and (2) “inflating actuals by measuring offices that are not live.” Id. Following this letter, Shah and Purdy discussed how to go about terminating Prowker. GX 846, 847.
These were not the only occasions where employees shared concerns about Outcome's business practices and Defendants ignored or dismissed the concerns, or sought to silence the employees. In late January 2017, Outcome employee Sahir Raoof met with the Nagaldinne to “formally lodge a complaint alleging willful misrepresentation of inventory data by ‘sales and management.'” GX 820. Agarwal shared this with Shah and Desai, including Raoof's report that “the entire [growth strategy] team is worried that we might be perpetuating fraud and some of them have gone to lawyers to ask for advi[c]e.” Id. Over the next few weeks, Shah discussed with Agarwal and Desai whether employees such as Raoof were “the right fit” and if they weren't, they should “quickly take decisive action” to rid Outcome of “toxic” personnel. GX 850, 865; see also GX 856.
In February 2017, Shah convened a company-wide town hall to address the recent departures, attributing the exits to “cultural differences” and inviting employees to meet with him one-on-one. GX 1945; Tr. 7436-39. Bridget O'Donnell was a newly hired employee working under Prowker who had noticed differences between the ROI results calculated by third-party vendor IMS and those reported to clients, which her team had shared with Prowker before his meeting with Shah and subsequent departure. GX 774, 774a; Tr. 7408, 7411, 7414, 7418-19. At the end of the town hall, O'Donnell approached Shah and relayed her concerns about Prowker's abrupt exit and that Outcome was “misleading clients based on the discrepancies” her team had found. Tr. 7440-43. O'Donnell testified that Shah responded that there were “operational inefficiencies that would explain the discrepancies,” id. 7444, and asked O'Donnell to email him the documents, which she declined to do fearing termination, id. 7449.
Another Outcome employee who reported to Prowker, Bryan Morgan, emailed Purdy and Desai two weeks later about his previously raised “concerns regarding how Outcome Health programs deploy vs how they are contracted, in addition to the glaring inconsistencies between [ROI] results presentations received from IMS and presentations that are ultimately delivered to clients.” GX 862; Tr. 4120-21. After forwarding the email to Shah, GX 862, Purdy responded to Morgan the next day that they had accepted his resignation, GX 863.
Shah and Purdy were acquitted on the wire fraud charge brought in connection with Purdy's forwarding of Morgan's email to Shah.
Purdy had heard about and disregarded employee concerns about Outcome's business practices almost a year prior. Between March and July 2016, several analysts working under Desai left Outcome in quick succession. Tr. 3961-66, 637080. Before leaving, the analysts completed exit questionnaires, sent initially to Williams, who forwarded them to Purdy. GX 580, 585, 586, 592. One analyst wrote that the “key qualities and skills” that Outcome should seek in her replacement were “[s]omeone for whom integrity is not a #1 concern; they should be willing to operate in ethical shades of gray,” adding that she “did not enjoy the ambiguity around what was considered the truth.” GX 580. Another analyst wrote that he began looking for a new job because of “[i]nconsistencies in what was promised and what was delivered” and the “[l]ack of sincerity and integrity from supervisors and leadership and ultimately being put in situations that felt morally compromising.” GX 586 (describing how he felt Outcome's “approach and actions” were “governed by . . . a need to churn profits often at the expense of ethics”). Another analyst quit because of “[e]thical issues with how [sales team] 1) reports ROI 2) fulfills contracts and 3) uses the analytics team to cover up these issues.” GX 592. Williams testified that he raised his concerns about what was being said in the exit interviews with Purdy, who responded “that the analysts were young and didn't necessarily understand the business model.” Tr. 6380-81.
In October 2017, the Wall Street Journal published an article reporting years-long fraud and unethical business practices at Outcome. Tr. 2459-60. Shah discussed the article at a company-wide townhall shortly thereafter. GX 1012. When asked about when “the allegations of [ROI report] manipulation [were] first raised directly to [him],” Shah claimed he had “learned about this specific concern . . . about a few- few weeks ago,” id., despite what Kazi, Prowker, Morgan, Raoof, and O'Donnell reported months prior.
Shah and Agarwal argue that the charged communications were not “for the purpose of executing” the alleged scheme. They highlight that they have found no case holding that such “internal corporate communications” satisfy the “for the purpose of executing” element of the mail and wire fraud statutes. They distinguish the communications at issue from “lulling” communications involving an alleged victim. R. 487 at 8-9. But the Seventh Circuit recently held, outside of the lulling context, that “[w]ire communications ‘that assist the defendant in avoiding detection may be sufficient to further a scheme.'” United States v. Lee, 77 F.4th 565, 572-73 (7th Cir. 2023) (quoting United States v. McGowan, 590 F.3d 446, 457 (7th Cir. 2009)). Shah and Agarwal's discussions about potentially firing employees who raised concerns served precisely that purpose. The backdrop of these communications in early 2017 was the effort at that time to secure hundreds of millions in capital funding. If the whistleblowers' concerns became public, that funding effort would undoubtedly have been at risk. Viewed in the light most favorable to the verdict, Shah and Agarwal's discussions about firing Kazi, Pierce, and Prowker were part and parcel of executing the scheme to defraud, or at the very least were “a step in the plot.” Sheneman, 682 F.3d at 629 (quoting Schmuck, 489 U.S. at 710-11). As such, there is sufficient evidence to sustain the convictions on Counts 14 through 19.
4. Defrauding Lenders and Investors (Counts 7, 9, 13, 22, 24, and 26)
Defendants argue that there is insufficient evidence that they knowingly participated in the scheme to defraud lenders and investors and intended to do so. The relevant counts are as follows:
- Count 7 charged Purdy with his email to James McHugh and others at JPMorgan with the borrower authorization letter on March 17, 2016 (GX 551).
- Count 9 charged Defendants with the closing of the $110 million loan on April 8, 2016.
- Count 13 charged Defendants with the closing of the $375 million loan to acquire AccentHealth on December 23, 2016.
- Count 22 charged Defendants with the investment from Goldman Sachs on March 1, 2017.
- Count 24 charged Defendants with the investment from CapitalG on March 20, 2017.
- Count 26 charged Defendants with the investment from Leerink on April 28, 2017.
The evidence showed that Defendants each knowingly and intentionally engaged in this part of the scheme, first by misleading the auditor, and then by using the auditor's opinions that Outcome's financial statements were free from material misstatement and other misrepresentations to obtain funding from lenders and investors.
By 2016, Defendants had started to look for outside investors, which required that a “Big Four” accounting firm audit Outcome's financial statements. Tr. 3977-78, 6886-87, 7767. Outcome hired Deloitte to audit its 2015 financial statements (“2015 audit”) in 2016, and its 2016 financial statements in 2017 (“2016 audit”). Id. 6893.
For both audits, Deloitte reviewed records provided by Outcome, interviewed Defendants and other employees, and used testing procedures to determine whether, in its professional opinion, Outcome's financial statements were free from material misstatement. Id. 6887-93. The evidence, viewed in the light most favorable to the verdict, demonstrated how Defendants misled Deloitte at each step of the way.
First, Defendants declined to say anything about overselling, under-delivering, or concealing under-deliveries during their “fraud-inquiry meetings” with Deloitte. The very purpose of those meetings was to inform Deloitte about what Defendants viewed “as risks to the financial statement, including fraud, and then to confirm or get any understanding as to what they may know about fraud, suspected fraud, alleged fraud, [and] noncompliance with laws and regulations.” Tr. 6978; see also GX 652. Additionally, before the audit closed, Deloitte checked back in with Purdy to ask whether there was anything else they should know about allegations of fraud, to which Purdy responded there was not. Tr. 7114-16. Indeed, the written engagement letters signed by Purdy provided that Deloitte expected fulsome and continuing disclosures of actual, suspected, or alleged fraud. GX 1932, 1933; Tr. 6984-85.
Along these lines, Deloitte expected Defendants to disclose that Outcome routinely: (1) sold advertising campaigns to pharma-clients based on projected inventory; (2) reassigned inventory between pharma-clients' campaigns; (3) inflated patient engagement metrics for tablets; and (4) created reports tracking the gap between contracted and installed inventory for campaigns. Tr. 6985, 6990, 7009. Deloitte also expected Defendants to disclose that numerous employees had either quit or been fired after raising allegations of fraud and that Desai had been accused of manipulating ROI reports. Id. 7108-7110, 7116-17. Yet Defendants did not tell Deloitte about these practices or about employees who shared their concerns about the practices, including Kazi, Prowker, Morgan, Raoof, and O'Donnell, all of whom raised their concerns while the 2016 audit was ongoing. GX 652; Tr. 7083-84, 7096, 7113, 7116, 7118, 7120.
Defendants also signed “management representation letters,” which Deloitte relied upon in issuing its opinions. Tr. 6984, 7004-06, 7121. The letters stated, contrary to the evidence discussed above, that Outcome had made available to Deloitte “all financial records and related data”; Defendants “ha[d] no knowledge of any fraud or suspected fraud affecting the company involving management”; and Defendants “ha[d] no knowledge of any allegations of fraud or suspected fraud affecting the company's financial statements.” GX 562, 910; Tr. 7008-10, 7123-25. Despite the fact that Deloitte had multiple interactions with Defendants, made numerous visits to Outcome's offices during the audits, and obtained contracts, campaign activity reports, and other documentation from Outcome, Deloitte had never heard of a “delta report” or “growth model.” Tr. 7008-10, 7125.
There was also evidence that Purdy knew the campaign activity reports sent to Deloitte were false. In the early stages of the 2015 audit, Desai emailed Purdy that Deloitte had asked for advertising contracts and related campaign activity reports to compare screens or offices where Outcome was playing advertisements against contractually-required numbers. Id. 3982-83, 7048-49. Desai forwarded Purdy an analyst's (Kathryn Choi) email identifying programs with deltas of 50% or more. Id. 3988-92; GX 530. Six days later, in response to Purdy's email asking if the two should have a phone call, Desai responded that they planned to “backfill” the campaign activity reports submitted to Deloitte with devices that were not playing advertisements to cover up the deltas. Tr. 4003-04; GX 530 (Desai explaining, “Between pulling in other installed scale, showing device assets (even those uninstalled/repair), we should be pretty good....”). Desai testified that he spoke with Purdy in the intervening period by phone about how much delta they could show on the reports and about backfilling, though the phone records did not reveal any call between the two during that time, Tr. 3997-4001, 5243, 5548; DX 3280 at 81; Tr. 5243. Desai and his team backfilled in the same exact way for the 2016 audit. Tr. 4005, 5549. Yet Deloitte never learned that the reports it received were false. Id. 6949, 7049.
Relying on Defendants' representations, Deloitte issued opinions that Outcome's 2015 and 2016 financial statements were free from material misstatement. GX 1298, 1301; Tr. 7012-15, 7127-29. Expert testimony revealed that in reality, Outcome's revenue for those years was overstated by $15 million and $29.8 million, respectively. Tr. 8540.
Outcome used those audits to secure an $110 million loan from JPMorgan and other lenders in April 2016 (Count 9), a $375 million loan from JPMorgan and other lenders in December 2016 to acquire Outcome's competitor AccentHealth (Count 13), and over $487 million from numerous investors between March and July 2017, including $100 million from Goldman Sachs (Count 22), $50 million from CapitalG (Count 24), and $15 million from Leerink (Count 26). GX 573, 698, 1081; Tr. 6695, 7920, 8642. The April 2016 loan could not close until the lenders received the audited financial statements. See GX 559 (March 2016 Voxer from Shah to Purdy noting how “central” getting the audit to the lenders was to the timing of the financing and directing Purdy to get it “wrapped up” by the next day). Further, the revenue numbers presented for the December 2016 loan mirrored the audited 2015 financial statements. GX 671, 1298. Additionally, testimony from lenders and investors made clear that the audits were important to them and they would have expected to know if Deloitte's opinions had been obtained by concealing material information about Outcome's contractual obligations or if revenue were overstated by millions. Tr. 6684-85, 6689, 6704-06, 7748-49, 7760, 7767, 7903-05, 8006, 8621-23, 8643-44.
Defendants had several opportunities to inform the lenders and investors of Outcome's fraudulent business practices, concealment from Deloitte, and the overstated revenue as part of the diligence process. Id. 7745, 7747-48, 7786-87, 7779-82. In March 2016, all three Defendants met with JPMorgan and other lenders and gave a detailed presentation about Outcome. GX 535. Purdy put together the presentation, seeking out Shah and Agarwal's feedback before the meeting. Id. At the meeting, Defendants did not say anything about over-selling, under-delivering, and concealing its gaps from clients. Instead, the presentation touted historical performance, rapid growth, and accelerating revenue. GX 535 at 17. Three days after the meeting, Purdy executed the borrower authorization letter, representing that Outcome's disclosures to the lenders “did not contain any untrue statement of material fact or omit to state a material fact necessary in order to make [the disclosures] not materially misleading[.]” GX 551 (Count 7). In November 2016, Shah and Purdy met with lenders and again presented on Outcome's strong growth and revenue. GX 671; Tr. 7786-87. But Shah and Purdy did not disclose that Outcome had obtained contractual renewals from pharma-clients by hiding delivery shortfalls, that it had conducted an internal audit revealing massive under-deliveries, or that numerous analysts had left that year citing ethical concerns.
During the capital raise in early 2017, Shah asked Desai about the downward modification of forecasted revenue for the first quarter, noting that he wasn't sure Outcome “can afford to have the revision. [Purdy] thinks it may violate a leverage covenant in the [December 2016] loan,” which would result in their “los[ing] the company to its lenders.” GX 714. Desai explained that the revenue losses were due to a “significant operational issue” in 2016 with the Entresto campaign, where there was a missed delivery by nearly 50% in the first six months, which resulted in poor ROI results and Outcome owing a sizable make good to the pharma-client. GX 715. Desai also identified other brands for which Outcome had missed its contractual ROI guarantees in 2016, requiring millions more in make goods and a lower revenue forecast for Q1 2017. Shah discussed the issue with Agarwal, raising the possibility of “severe repercussions.” GX 736. Shah and Purdy discussed the issue as well. Id.
Yet, Defendants never raised these issues with Deloitte as it conducted the 2016 audit or during multiple meetings with the investors they met with over the same time period. GX 729, 777; Tr. 6694-95, 6705, 6713, 7917, 7923, 8624. Instead, Shah and Purdy reiterated that Outcome had never missed an ROI guarantee, despite knowing that was not the case, and gave alternative explanations for Outcome's revenue miss. GX 697A, 715, 736, 757, 932; Tr. 7914-18, 8628-29.
Additionally, both Shah and Agarwal stood to benefit from the loans and investments. Shah and Agarwal received dividends of $30.2 million and $7.5 million, respectively, with the April 2016 loan and a $225 million distribution from the 2017 capital raise. GX 571, 572, 1033, 1071, 1078; Tr. 6673-74. Further, Defendants each held an ownership stake in Outcome, while Desai did not. GX 1064, 1941. This evidence, taken altogether and in the verdict's favor, is sufficient to show Defendants' knowing participation in defrauding lenders and investors and intent to do so.
Defendants challenge the sufficiency of the evidence on these counts in several ways. First, they reiterate their argument that it was Desai's fraud, which was entirely hidden from them, and they could not disclose information they did not know. Relatedly, they argue that there is no evidence that they knew the revenue was inflated or the ROI numbers were inaccurate. For the reasons previously discussed, those arguments do not hold water.
Defendants similarly argue that they did not understand the parade of employees to be raising concerns about fraud but merely operational problems. But the reasonable inference-drawn in the government's favor-from the volume and content of the concerns raised is that Defendants were aware of, at a minimum, suspected or alleged fraud. E.g., GX 760 (relaying Kazi's statement that “it is fraud”); GX 1911 (relaying Prowker's allegation of “fraudulent” conduct); GX 820 (relaying Raoof's report that “the entire [growth strategy] team is worried that we might be perpetuating fraud”). Defendants highlight the sub-certifications by Outcome managers who did not raise material concerns. DX 10714-54. But the fact that others certified that they were not aware of any fraud does not change the fact that Defendants were each independently aware of concerns raised by employees about the integrity of Outcome's practices and decided not to raise them to Deloitte notwithstanding several opportunities to do so.
Shah and Purdy further challenge the contention that they “misrepresented” that Outcome had never missed an ROI guarantee. Shah says that Outcome's presentation decks “tell the real story”: that Outcome had never paid out of pocket for a missed ROI guarantee. R. 489 at 9 (citing GX 729, 777). Defendants' argument that such hair-splitting interpretations were accepted by investors risking tens or hundreds of millions of dollars is simply not plausible and is belied by the investor testimony itself. First, Sturdy and Cozzens recalled the verbal representations made by Shah and Purdy that Outcome had never missed an ROI guarantee. Tr. 7914-18, 8628-29, 8717. Purdy calls out Sturdy's and Cozzens' imprecise recollections of who made the statements and who was in the room at the time and how the audited financial statements that the investors received differentiated between a make good and a cash refund. See id. 8755; GX 723, 917. However, Cozzens testified that he mostly spoke about such matters with Shah and Purdy, and Sturdy's testified that she recalled asking Shah and Purdy specifically whether Outcome had “missed an ROI guarantee” and there were no discussions of make goods before the investment. Tr. 7616, 8716-17. Drawing all reasonable inferences in the government's favor, the evidence supports that Shah and Purdy misrepresented that Outcome had never missed an ROI guarantee.
Shah also claims that the decision to hire an auditor and give them “full access” to Outcome's information is counter to an intent to defraud. He similarly points out that one whistleblower's complaint that Outcome “sold digital advertising inventory that it did not have to a [pharma-client]” was disclosed to investors, DX 4737, and that he and Purdy disclosed the revenue shortfall to Leerink, DX 12255. But the voluntary nature of the audit and those select disclosures do not negate testimony and other evidence, which must be viewed in the light most favorable to the verdict, that Defendants did not tell Deloitte, lenders, or investors about overselling inventory, under-deliveries, falsified affidavits, altered ROI studies, and employees' reports of fraud concerns. Half-truths are still false.
Agarwal contends that the jury's verdict on these counts was grounded in speculation. As to the loans, Agarwal highlights testimony that she “had no role in providing financial information” to McHugh; that she was not a presenter at the March 2016 presentation; and that she was not present for the November 2016 presentations. Tr. 7781-82, 7842-45; see also GX 671. As to the capital raise, she emphasizes that there is no evidence that she provided financial information in connection with the investments; that she never met with Eberts; that her conversations with Sturdy were “mostly around product, content, and culture”; and that she had limited interactions with Cozzens about the company and its mission. Tr. 6859-60, 7894, 8612-13. Critically, however, Agarwal provided information to Deloitte during the audits, through fraud inquiry meetings and management representation letters. The lenders and investors relied on Deloitte's sign-off. Additionally, while Agarwal's communications with lenders and investors were less frequent than Shah or Purdy, they did occur, and Agarwal did not share the truth about Outcome's practices or reports of fraud.
Purdy argues that the only evidence of his knowledge of the falsity of Outcome's financial statements was Desai's testimony about their phone call. In so arguing, Purdy asks this Court to find that testimony incredible as a matter of law. He argues that Desai's testimony about when the call took place cannot be reconciled with the evidence that Purdy was vacationing in Mexico at that time, the lack of any phone records showing a call between the two during the period, the timing and content of Purdy's email in response, GX 530, and Desai's emails with Choi, DX 3680, 3695.
But Desai's testimony about the phone call is not “contrary to the laws of nature or so internally inconsistent or implausible on its face that no reasonable factfinder would credit it.” United States v. Faulkner, 885 F.3d 488, 493 (7th Cir. 2018); see also United States v. Sophie, 900 F.2d 1064, 1079 (7th Cir. 1990) (referring to testimony that is “inherently unbelievable in the sense that it contradicted the laws of nature or other indisputably true evidence”). First, the fact that Desai's phone records also do not include a call during the time frame does not rule out the possibility that the call occurred. See DX 3649. The evidence showed that Purdy's phone records also omitted a call that indisputably took place between Purdy and Deloitte while he was in Mexico during the same period. DX 3280, 5189.1; Tr. 553435, 7003-04. And Purdy's email in response does not rule out the possibility that he and Desai could have spoken over the phone in the intervening six days. For this reason, the jury could have disregarded the phone records as a credible source of information. Second and as discussed, the phone call was not the only evidence of Purdy's misrepresentations to the auditor, investors, and lenders. The jury could have set aside the issue of the phone call and found that Purdy signed off on the plan to send false activity reports to Deloitte based on the fact that Desai told him explicitly what he and Choi planned to do, and Purdy did not direct them to do otherwise. See GX 530, Tr. 5542-43, 5545.
Considering all of the evidence in the light most favorable to the verdict and drawing all reasonable inferences for the government, the evidence is sufficient to sustain the convictions on the counts relating to defrauding investors and lenders.
B. False Statement to a Financial Institution (Count 8)
The jury convicted Purdy of making a false statement to a financial institution (18 U.S.C. § 1014). In order to prove him guilty of this offense, the government had to prove beyond a reasonable doubt that (1) he made a false statement to a financial institution; (2) knowing the statement was not true when it was made; (3) with the intent to influence the financial institution to take some action; and (4) the financial institution was FDIC insured. R. 444 at 41 (Instruction No. 38); see also Seventh Circuit Pattern Criminal Jury Instructions at 453. Count 8 charged Purdy with the transmission of the audited financial statements for 2015 to JPMorgan on March 23, 2016. As discussed previously, the evidence, considered in the light most favorable to the verdict, showed that Purdy and the other Defendants deceived Deloitte in obtaining the audited 2015 financial statements, that the lenders would not have authorized the April 2016 loan without the audited financials, and that Shah instructed Purdy to get the audit wrapped up as quickly as possible because it was critical to securing the loan. That evidence, along with the stipulation that JPMorgan was FDIC insured, is sufficient to sustain the guilty verdict on Count 8.
C. Money Laundering (Counts 10 and 12)
The jury convicted Shah of money laundering (18 U.S.C. § 1957) in connection with the payments made for his residence and flight transportation. See GX 571, 577, 587, 1242. In order to prove Shah guilty of this offense, the government had to prove beyond a reasonable doubt that he (1) engaged or attempted to engage in a monetary transaction; (2) knowing the transaction involved criminally derived property; (3) the property had a value of greater than $10,000; (4) the property was derived from bank fraud; and (5) the transaction occurred in the United States. R. 444 at 38 (Instruction No. 35); see also Seventh Circuit Pattern Criminal Jury Instructions at 807. There is no dispute that the charged transactions involved proceeds from the April 2016 bank loan, were initiated by Shah, and occurred in the United States for over $10,000. Shah's only challenge is that there was insufficient evidence that he knowingly and intentionally engaged in any fraud. As discussed in detail in the preceding sections, the Court finds otherwise. The evidence is sufficient to sustain the guilty verdict on these counts.
II. Motions for New Trial
Defendants argue that even if the Court denies their motions for judgment of acquittal, they are entitled to a new trial. A district court may grant a new trial “if the interest of justice so requires.” Fed. R. Crim. P. 33(a). Unlike a motion for acquittal under Rule 29, in considering a motion for a new trial under Rule 33, the Court does not review the evidence in the light most favorable to the government. United States v. Washington, 184 F.3d 653, 657 (7th Cir. 1999). Instead, courts “may reweigh the evidence, taking into account the credibility of the witnesses.” Id. at 658. Yet despite this more lenient standard, granting a new trial is “reserved for only the most extreme cases.” Coscia, 4 F.4th at 465; see also United States v. Santos, 20 F.3d 280, 285 (7th Cir. 1994) (“A jury verdict in a criminal case is not to be overturned lightly, and therefore a Rule 33 motion is not to be granted lightly.”). “A new trial should be granted only if the evidence preponderates heavily against the verdict, such that it would be a miscarriage of justice to let the verdict stand.” United States v. Friedman, 971 F.3d 700, 713 (7th Cir. 2020) (citations omitted).
This is not one of the extreme cases that warrants a new trial. After viewing the trial, carefully observing the testimony and demeanor of the witnesses, and thoroughly reviewing the extensive record, the Court does not have “strong doubt” as to Defendants' guilt on the charged offenses. See Washington, 184 F.3d at 658. Given all of the incriminating testimony and exhibits described above that illustrate Defendants' knowledge and intent, it is not a manifest injustice to let the verdict stand.
While Shah and Agarwal nominally move in the alternative for a new trial under Rule 33, they offer no new argument beyond that raised for acquittal. Purdy specifically argues that even if Desai's testimony in connection with his phone call with Purdy about the information to be shared with Deloitte does not require a judgment of acquittal, it leaves strong doubt as to Purdy's guilt. But Desai's testimony over the course of 11 days was incredibly detailed and corroborated by an extensive documentary record. With regard to the phone call specifically, Desai maintained his clear recollection of his conversation with Purdy. E.g., Tr. 5548. And in any case, Desai's testimony did not stand alone as evidence of Purdy's knowing participation in the scheme and intent to defraud pharma-clients, investors, and lenders. For those reasons, the Court denies Defendants' Rule 33 motions for a new trial.
Conclusion
For the foregoing reasons, the Court denies the motions for judgment of acquittal and the motions for a new trial.