UNPUBLISHED
UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT
No. 23-2027
REDDY VIJAY ANNAPPAREDDY, Plaintiff – Appellant,
v.
UNITED STATES OF AMERICA, Defendant – Appellee,
and
MAURA LATING; ROBERT MOSLEY; PAM ARNOLD; JAMES P. RYAN; SANDRA WILKINSON; CATHERINE SCHUSTER PASCALE; STEVEN CAPOBIANCO,
Defendants.
Appeal from the United States District Court for the District of Maryland, at Baltimore. Joseph F. Anderson, Jr., Senior District Judge. (1:18-cv-03012-JFA)
Argued: October 29, 2024 Decided: January 28, 2025
Before GREGORY and HARRIS, Circuit Judges, and Louise W. FLANAGAN, United States District Judge for the Eastern District of North Carolina, sitting by designation.
Affirmed by unpublished opinion. Judge Harris wrote the opinion, in which Judge Gregory and Judge Flanagan joined.
2
ARGUED: Joshua Daniel Greenberg, THE JOSH GREENBERG LAW FIRM PLLC, Washington, D.C., for Appellant. Molissa Heather Farber, OFFICE OF THE UNITED STATES ATTORNEY, Baltimore, Maryland, for Appellee. ON BRIEF: Erek L. Barron, United States Attorney, OFFICE OF THE UNITED STATES ATTORNEY, Baltimore, Maryland, for Appellee.
Unpublished opinions are not binding precedent in this circuit.
3
PAMELA HARRIS, Circuit Judge:
After federal criminal charges against him were dismissed, Reddy Annappareddy sued the United States for malicious prosecution under the Federal Tort Claims Act. The district court held a three-week bench trial and then ruled against Annappareddy on two alternative grounds: first, that he failed to show the proceedings against him were instituted without probable cause; and second, that he had not shown a malicious motive for the prosecution. We agree on both points and affirm the district court’s judgment.
I.
This case began with the government’s criminal prosecution of Reddy Annappareddy for healthcare fraud. After his conviction was vacated and the charges against him dismissed, Annappareddy sued the United States and multiple state and federal officials involved in his prosecution. In a previous opinion, we held that most of Annappareddy’s claims should be dismissed. See Annappareddy v. Pascale, 996 F.3d 120 (4th Cir. 2021) (Annappareddy I).
But we have not yet considered Annappareddy’s claim against the United States under the Federal Tort Claims Act (“FTCA”), alleging that two federal agents, Maura Lating and Robert Mosley, knowingly relied on false and unreliable information in obtaining a search warrant and indictment. The facts related to the agents’ investigation of Annappareddy are recounted in detail in our prior opinion, see id. at 127–29, and we outline them just briefly here. We then turn to Annappareddy’s criminal proceedings and his subsequent civil suit, and finally to the district court ruling now on appeal.
4
A.
Plaintiff Annappareddy founded and owned Pharmacare, a chain of pharmacies located in Maryland and other nearby states. State and federal officials began investigating Pharmacare in 2012, when its former Chief Operating Officer, Dennis Tokofsky, reported the company to Maryland authorities, alleging that it regularly billed Medicaid for automatically refilled prescriptions that were never picked up or delivered to patients. A pharmacist at one of Pharmacare’s stores, Lisa Ridolfi, echoed Tokofsky’s account and became a confidential informant, helping the team to gather evidence.
FBI Special Agent Maura Lating and U.S. Department of Health and Human Services Special Agent Robert Mosley – the targets of Annappareddy’s FTCA claim – were members of the investigative team. Together, the team spent approximately 13 months investigating and corroborating Tokofsky’s initial allegations before seeking a search warrant or indictment. Investigators conducted multiple interviews with current and former Pharmacare employees, who validated the overall billing-but-not-filling scheme. They also interviewed patients, who likewise confirmed Tokofsky’s allegations, reporting that they were billed for prescriptions that had been automatically refilled but never delivered. Additional evidence was discovered through undercover operations, surveillance, and trash pulls.
Agent Mosley also worked with a Medicare drug integrity contractor (“MEDIC”)
to prepare a statistical analysis of Pharmacare’s invoices and inventory. The MEDIC analysis purported to find “shortages” of multiple medications – that is, that Pharmacare had insufficient inventory to fill prescriptions for which it billed and was paid. This
5
calculation was designed to estimate the financial loss incurred by government insurance programs as a result of the alleged fraud.
Ultimately, the team sought a search warrant for a number of Pharmacare locations.
Agent Lating signed the 40-page supporting affidavit, which described in detail the evidence obtained from two confidential informants – Tokofsky and Ridolfi – and from the team’s corroborating investigation. It also included the MEDIC shortage calculations. Based on this affidavit (the “Lating Affidavit”), a federal magistrate judge found probable cause to search six locations. On the same day, Agent Mosley used the Lating Affidavit as an outline for his grand jury testimony, which led to the indictment of Annappareddy and two of his employees on charges of healthcare fraud and aggravated identity theft.
B.
Because Annappareddy’s claim is that Lating and Mosley knowingly relied on false information in obtaining the search warrant and indictment against him, events that came later – after presentation of the Lating Affidavit to the magistrate judge and grand jury – are not directly relevant here. But for context, we summarize the rest of the events that led to this appeal.
Following a superseding indictment and jury trial, Annappareddy was convicted of healthcare fraud and identity theft. That conviction was vacated, however, when Annappareddy moved for a new trial, and discovery produced evidence of significant flaws in an updated MEDIC analysis used against him at trial. As a result of that disclosure, the government joined Annappareddy’s request for a new trial, which was granted by the district court. And then, after securing a new superseding indictment, the government
6
disclosed that it had destroyed three boxes of relevant documents – mistakenly, according to the government, but intentionally in Annappareddy’s view. Annappareddy moved to dismiss all charges with prejudice, and the district court granted the motion.
Annappareddy then filed the 25-count civil lawsuit that gives rise to this appeal, naming as defendants seven individual state and federal officers and the United States. Most of those claims have been resolved already. In our first review of this case, on interlocutory appeal, we affirmed the dismissal of Annappareddy’s claims against the individual federal officers, agreeing with the district court that a Bivens cause of action could not be extended into this new context. See Annappareddy I, 996 F.3d at 132–38. We also held that Annappareddy’s claims against a state prosecutor were barred by prosecutorial immunity. Id. at 138–42. But Annappareddy’s FTCA claims against the United States were still pending, see id. at 130 n.3, and we left that issue for further proceedings on remand. 1 C.
That brings us to the trial and ruling now before us on appeal. The FTCA allows for the recovery of damages against the United States for a malicious prosecution claim that involves the conduct of “investigative or law enforcement officers . . . who [are] empowered by law to execute searches, to seize evidence, or to make arrests for violations of Federal law.” 28 U.S.C. § 2680(h). Annappareddy’s FTCA claim accordingly focuses
1
There also remained state-law claims against one state investigator. See Annappareddy I, 996 F.3d at 130 n.3. The district court separated out those claims from the FTCA claim against the United States, and they are not relevant to this appeal.
7
on the actions of two federal agents, Lating and Mosley, alleging that they engaged in malicious prosecution when they used false and unreliable information to obtain both his indictment and the search warrants that effectively shuttered his pharmacies. 2 That FTCA claim is governed by state law, see 28 U.S.C. § 1346(b)(1) – here, by the law of Maryland. To make out a claim of malicious prosecution under Maryland law, Annappareddy was required to prove four elements: (1) a criminal proceeding instituted against him; (2) without probable cause; (3) with malice, or with a motive other than to bring an offender to justice; and (4) termination of the proceeding in his favor. Heron v. Strader, 761 A.2d 56, 59 (Md. 2000). The parties agreed that Annappareddy could satisfy the first and fourth elements, but contested the second and third – lack of probable cause and malice. See Annappareddy v. Lating, No. 1:18-cv-03012, 2023 WL 6538392, at *24 (D. Md. Aug. 3, 2023).
The district court thus proceeded to a lengthy bench trial. Over the course of ten days, the court heard approximately 40 hours of testimony from 19 witnesses, including Annappareddy, and reviewed approximately 200 exhibits. It then set out its findings of fact and conclusions of law in a detailed and thoroughly reasoned 67-page opinion. The court gave special care to its credibility determinations, addressing the testimony of each
2
Annappareddy initially pressed five FTCA counts against the United States. The district court granted summary judgment to the United States on three of those counts, see Annappareddy v. Lating, No. 1:18-cv-03012, 2023 WL 2540306, at *18 (D. Md. Mar. 16, 2023), and Annappareddy did not appeal that judgment. The district court handled together the remaining two counts – malicious prosecution in obtaining the search warrant and malicious prosecution in the indictment – and we do the same.
8
witness in turn to provide “greater insight,” id. at *3, into its bottom-line finding: that the government’s witnesses, including Lating and Mosley, were credible, see id. at *15–16, while Annappareddy was not, see id. at *22–23 (explaining that the “court had difficulty giving Annappareddy’s testimony credence at any point,” and discussing one particularly “concerning” inaccuracy).
Based on its credibility determinations and factual findings, the district court held, first, that Annappareddy had failed to prove that the proceedings against him were not supported by probable cause. The crux of the issue here, as the district court explained, was the Lating Affidavit, on which both the search warrants and the indictment rested. See id. at *5, *23 & n.31. Applying Maryland law, the district court considered the sufficiency of that affidavit using a Franks framework, under which Annappareddy could challenge probable cause only by showing both culpability and materiality: that the agents intentionally or recklessly included false information in the affidavit; and that without the false information, there would not be enough for probable cause. Id. at *24–25 (discussing Franks v. Delaware, 438 U.S. 154, 155–56 (1978)).
Under that standard, the district court emphasized, what mattered was what Lating and Mosley knew or had reason to believe at the time the Lating Affidavit was submitted. And there was no credible evidence, the court found, that either Lating or Mosley had any reason to believe that the Lating Affidavit included false information. See id. at *15–16. To the contrary: The “mountain of corroboration” unearthed by the team’s investigation meant that Lating and Mosley had no reason to doubt the veracity of their confidential informants or the accuracy of the affidavit. Id. at *26. The affidavit did include, the court
9
recognized, an early version of the flawed MEDIC invoice analysis that led to the vacatur of Annappareddy’s conviction, which contained some “duplication errors” and did not give a “complete picture of Pharmacare’s actual operation.” Id. at *27. But again, there was “no indication” that the agents were aware of those errors prior to Annappareddy’s indictment, and in any event, they were immaterial to a probable cause finding. Id.
The court also held, in the alternative, that Annappareddy could not prevail on his malicious prosecution claim because he could not prove the requisite malice. Id. at *29. The court recognized that under Maryland law, malice may be inferred from a lack of probable cause. See id. at *29 n.37 (citing Wesko v. G.E.M., Inc., 321 A.2d 529, 532 (Md. 1974)). But here, the court explained, no such inference would be appropriate because there was no evidence that Mosley or Lating – or any agent – had any ill-will toward Annappareddy or “ever acted for personal gain or for any reason other than to bring about justice.” Id. at *29. Instead, Mosley and Lating, along with the rest of their team, initiated an investigation based on reports of fraudulent activity, took normal investigative steps, and made “painstaking efforts to verify, corroborate, and gather evidence of fraud” – leaving them, in the end, with no doubt that probable cause existed. Id. Annappareddy’s claims of malice were wholly unsupported by the evidence, the court concluded, and for this reason, too, his “claims for malicious prosecution must fail.” Id.
The district court entered final judgment in favor of the United States, dismissing Annappareddy’s FTCA claim with prejudice. Annappareddy timely appealed.
10
II.
We review a judgment following a bench trial under a mixed standard of review.
Roanoke Cement Co. v. Falk Corp., 413 F.3d 431, 433 (4th Cir. 2005). Conclusions of law are reviewed de novo. Id. But we may reverse a finding of fact only if it is “clearly erroneous.” Id.
That deferential standard of review is important to this appeal, in which Annappareddy primarily disputes the district court’s weighing of the evidence at trial. But even if we might have “weighed the evidence differently,” we will not reverse the district court so long as its “account of the evidence is plausible” on the record as a whole. Anderson v. City of Bessemer City, N.C., 470 U.S. 564, 574 (1985). And where, as here, a district court’s findings rely in part on an assessment of witness credibility, we give those findings even greater deference, recognizing that a district court has the critical benefit of observing the witnesses in person. See U.S. Fire Ins. Co. v. Allied Towing Corp., 966 F.2d 820, 824 (4th Cir. 1992).
A.
We begin with the district court’s holding that Annappareddy could not satisfy the probable cause element of his malicious prosecution claim. On appeal, Annappareddy does not challenge the legal standard applied by the district court, incorporating the Franks framework and requiring him to show that Lating and Mosley intentionally or recklessly included false information in the Lating Affidavit that was material to the probable cause determination. Annappareddy, 2023 WL 6538392, at *24. Instead, Annappareddy insists that he did make that showing at trial – that he proved by a preponderance of the evidence
11
that Lating and Mosley intentionally or recklessly put false and material information in the Lating Affidavit – and that the district court erred in finding to the contrary. We disagree and discern no clear error in the district court’s findings.
The district court carefully explained its analysis of the evidence and its factual findings on this matter, and we summarize only briefly here. Annappareddy’s primary argument was that the agents intentionally or recklessly included in the affidavit the accounts of two confidential informants – former COO Tokofsky and pharmacist Ridolfi – who they knew or should have known were biased and unreliable. See id. at *5–8, *26. But the district court, as noted above, found Lating and Mosley to be credible witnesses, and it credited their testimony that they believed Tokofsky and Ridolfi were reliable. Id. at *26. And, importantly, the district court found that belief was entirely reasonable, given the “mountain” of evidence corroborating the informants’ accounts. Id. The district court carefully reviewed the investigators’ “continuous[] work[] to corroborate the information” provided by Tokofsky and Ridolfi “through a variety of sources” – physical evidence and photographs provided by Ridolfi to back up her account; interviews with Pharmacare patients, some of whom recounted experiences “indicative of the exact scenarios described by Tokofsky and Ridolfi;” interviews with Pharmacare employees, who also gave accounts consistent with Ridolfi and Tokofsky’s testimony; and undercover operations using fake patient profiles. Id. at *7. “All told,” the district court found, “the multitude of corroborating information garnered from a variety of unrelated sources more than justified
12
[the] agents’ beliefs that Tokofsky and Ridolfi were reliable informants providing credible information.” Id. at *8. We see no clear error in that finding.
Annappareddy also tried to undermine the Lating Affidavit by arguing that Lating and Mosley recklessly or intentionally included false results from a preliminary MEDIC invoice analysis. The particulars of Annappareddy’s concerns regarding the MEDIC analysis – an alleged failure to account for transfers across stores, plus instances of “double counting” billings for prescriptions with secondary insurers – are complex and well explained in the district court’s opinion. See id. at *8–12. For present purposes, it is enough that the district court agreed – as did the government – that the MEDIC analysis contained some “inaccuracies” and “duplication errors.” Id. at *10, *27. But the court also found that neither Mosley, who worked with MEDIC, or Lating knew or had reason to know of those errors at the time the Lating Affidavit was submitted. Id. at *11, *27. We owe substantial deference to the district court’s decision to credit Mosley’s testimony – that he believed the information in the MEDIC analysis was accurate, id. at 15 – and also to credit corroborating witness testimony that consistently described Mosley as acting only as a “logistical liaison” to MEDIC, without the expertise to perform a technical analysis of his own, id. at *11. We see no clear error in the district court’s finding that Annappareddy failed to show that Lating or Mosley doubted or had reason to doubt the accuracy of the MEDIC analysis in the Lating Affidavit.
The district court also found, in the alternative, that Annappareddy had not proven that any MEDIC errors in the Lating Affidavit were material to the probable cause determination. Id. at *27. Again, we see no error in that finding. Even with the corrections
13
deemed necessary by Annappareddy and his expert, the district court explained, the invoice analysis still would show a shortage for several medications – a significant gap between inventory and billings that was indicative of fraud and supportive of probable cause. Id. And more fundamentally, the entirety of the MEDIC analysis could be removed from the Lating Affidavit without “consequence” for the probable cause determination because the MEDIC data was used to estimate the scale of the fraud and the amount of loss – and neither is an element of the offenses for which Annappareddy was prosecuted. Id. All the government needed to show probable cause, the court explained, was the multitude of evidence it gathered in corroborating its informants’ allegations – witness accounts, photos, claims data, and the like – making any reliance on the MEDIC invoice analysis immaterial. Id.
Finally, in one sentence in the argument section of his brief, Annappareddy claims that the Lating Affidavit falsely stated that pharmacies are required to reverse the billing for a prescription that is not picked up or delivered within 14 days, when in fact the law allowed for a 60-day window. We assume without deciding that this argument is preserved. But see Grayson O Co. v. Agadir Int’l LLC, 856 F.3d 307, 316 (4th Cir. 2017) (explaining that a party waives an argument by failing to “develop” it in a brief, even if the brief takes a “passing shot” at the issue). The district court rejected it on two grounds: First, the affidavit’s reference to a 14-day standard is not “false” because there is in fact a 14-day industry standard for reversing billings on old prescriptions, and the affidavit does not present that standard as a legal requirement. Annappareddy, 2023 WL 6538392, at *26. And second, references to a 14-day standard were in any event immaterial, as the
14
Lating Affidavit also discussed ample evidence of undelivered prescriptions older than the legally required 60-day reversal period. Id. at *27. We see no ground for reversal of those findings.
In sum, having considered Annappareddy’s arguments 3 and the full trial record, we find no error in the district court’s conclusion that Annappareddy failed to prove the probable cause element of his malicious prosecution claim.
B.
The district court also held that Annappareddy could not make out a malicious prosecution claim for a second, independent reason: he could not prove malice. Even if, that is, Lating and Mosley had initiated Annappareddy’s prosecution without probable cause, Annappareddy had not presented any evidence that they acted with a motive “other than that of bringing an offender to justice.” Annappareddy, 2023 WL 6538392, at *29 (citing Montgomery Ward v. Wilson, 664 A.2d 916, 922 (Md. 1995)). Again, we have considered Annappareddy’s arguments and find no basis for disturbing the district court’s ruling.
Under Maryland law, malice is a separate and independent element of the tort of malicious prosecution. To prove malice, Annappareddy must show that Lating and Mosley
Annappareddy also suggests that Lating intentionally or recklessly included in her
3
search warrant application a newer, closed-door pharmacy, CareMerica, as to which there was not probable cause. But as the district court explained, CareMerica opened at the site of what was formerly a Pharmacare store, and other Pharmacare stores transferred their prescriptions to CareMerica upon opening. Annappareddy, 2023 WL 6538392, at *14. We agree with the district court and the government that a new name on a building does not affect the probable cause analysis.
15
had a “wrongful or improper motive in initiating legal proceedings,” meaning some motive other than “bringing the party to justice.” Montgomery Ward, 664 A.2d at 924 (internal citation omitted). As the district court explained, Annappareddy attempted to make that showing at trial. He suggested that Lating and Mosley initiated his prosecution to win an award – but given the complete absence of evidence of any such “mythical award,” the district court rejected that theory as “completely manufactured.” Annappareddy, 2023 WL 6538392, at *29. Nor, the district court found, was there any evidence that Lating or Mosley knew of or disliked Annappareddy before the investigation, or had any reason to engage in an elaborate conspiracy to fabricate evidence, falsify data analyses, and lie to federal prosecutors. Id. Instead, the evidence showed that Lating and Mosley acted only “in the pursuit of bringing an individual they had no prior knowledge of to justice for his role in [a] reported fraud.” Id. Based on the trial record, this finding of an absence of malice was not clearly erroneous.
We also reject Annappareddy’s argument that the district court erred by not inferring malice from a lack of probable cause – a lack the court was assuming only for the purpose of its alternative holding. “While malice can be inferred from want of probable cause, it is no more than a permissible inference[.]” Wesko, 321 A.2d at 532. The existence of malice is ultimately a fact question for the trier of fact to decide. Banks v. Montgomery Ward & Co., 128 A.2d 600, 606 (Md. 1957); Torsch v. Dell, 41 A. 903, 905 (Md. 1898). In light of the evidence presented at trial, the district court did not err in holding, in the alternative, that it would decline to infer malice in this case even if probable cause were lacking.
16
III.
The evidence that Annappareddy presented at trial failed to prove two necessary elements of his malicious prosecution claim. We therefore affirm the district court’s entry of judgment in favor of the defendant, the United States.
AFFIRMED