Daisy Miller v. United States

Opinion

[DO NOT PUBLISH]

In the

United States Court of Appeals For the Eleventh Circuit

____________________

No. 20-11988

____________________

DAISY MILLER, Petitioner-Appellant,

versus UNITED STATES OF AMERICA,

Respondent-Appellee.

____________________

Appeal from the United States District Court for the Southern District of Florida D.C. Docket Nos. 1:16-cv-21090-JEM, 1:12-cr-20757-JEM-2

____________________

2

2 Opinion of the Court 20-11988

Before WILSON, NEWSOM, and LAGOA, Circuit Judges. LAGOA, Circuit Judge:

After a five-week trial, a jury found Daisy Miller guilty on various counts relating to Medicare fraud. During the trial, the jury heard from many witnesses and was presented with many documents corroborating the government’s theory of the case and discrediting Miller’s testimony. Following our affirmance of her conviction in United States v. Kallen-Zury (Kallen-Zury I), 629 F. App’x 894 (11th Cir. 2015), Miller moved to vacate her conviction under 28 U.S.C. § 2255. The district court denied the motion, and Miller now appeals that denial. On appeal, she argues that her trial counsel should have called several witnesses in her defense, that counsel ’s failure to do so constituted ineffective assistance of counsel, and that we should vacate her conviction and grant her a new trial.

For the reasons explained below, and with the benefit of oral argument, we conclude that trial counsel’s decision not to call these witnesses did not prejudice Miller or constitute deficient performance . Accordingly, we affirm.

I. FACTUAL AND PROCEDURAL BACKGROUND A. Underlying Criminal Case This is the second time Miller has come before this Court regarding her trial. We outlined the facts of this case in our previous decision affirming Miller and her co-defendants’ convictions in their direct appeal, see Kallen-Zury I, 629 F. App’x 894, as well as in

3

23-10182 Opinion of the Court 3

our decision affirming the district court’s denial of Karen Kallen- Zury’s—Miller’s co-defendant—motion for a new trial, see United States v. Kallen-Zury (Kallen-Zury II), 710 F. App’x 365 (11th Cir. 2017).

On October 2, 2012, a federal grand jury in the Southern District of Florida returned an indictment charging Miller with the following offenses: conspiracy to commit health care fraud and wire fraud, in violation of 18 U.S.C. § 1349 (Count 1); wire fraud, in violation of 18 U.S.C. §§ 1343 and 2 (Counts 2 to 6); health care fraud, in violation of 18 U.S.C. §§ 1347 and 2 (Counts 7 and 8); and conspiracy to defraud the United States and to pay and receive health care kickbacks, in violation of 18 U.S.C. § 371 (Count 9). The indictment also charged her co-defendants Karen Kallen-Zury, Christian Coloma, Michele Petrie, and Gloria Himmons with the same or related conspiracy, fraud, and kickback offenses. On November 2, 2012, Omar Malone was appointed pursuant to the Criminal Justice Act (the “CJA”) to represent Miller, who elected to proceed to trial.

At trial, the government established that Hollywood Pavilion (“HP”) is “a mental health facility that included both inpatient and outpatient treatment programs.” Kallen-Zury I, 629 F. App’x at 897. Miller was a licensed clinical social worker who began working for HP in 2002. At all relevant times, she was the clinical director for HP’s inpatient facility and ran its day-to-day operations. Id. Miller worked closely with Chris Gabel, HP’s Chief Operating Officer , and Dr. Alan Gumer, HP’s psychiatric medical director. Id.

4

4 Opinion of the Court 20-11988

During Miller’s tenure at HP, the facility paid recruiters to recruit patients on its behalf and received Medicare reimbursements for those patients. Id. This practice, however, is illegal. The government’s theory of the case was that Miller and her co-defendants conspired to defraud the United States by creating a health care kickback scheme through Medicare reimbursements and that Miller participated in the scheme in order to keep her job, title, salary , and status.

“The backbone of the government’s case was the testimony of several patient recruiters”—Keith Humes, Jean Luc Veraguas, Mathis Moore, Curtis Gates, and Gloria Himmons—“who pleaded guilty to Medicare fraud related to HP and other facilities.” Id. “These recruiters would find patients from as far away as Maryland and would pay to have the patients ride buses down to HP in Hollywood , Florida.” Kallen-Zury II, 710 F. App’x at 367. “Most of the[se] patients were drug addicts who did not need the psychiatric services offered at HP.” Id. Therefore, “the conspirators often falsified the patients’ records to reflect serious psychiatric problems or told the patients to claim psychiatric issues upon admission” to HP. Id. HP would then pay the recruiter for each patient the recruiter sent to its facility. See id. at 367–68. Additionally, HP only admitted “patients who had enough days on their Medicare plans to have their treatment periods paid for by the government.” Kallen-Zury I, 629 F. App’x at 897. When the Medicare money ran out for a patient, the facility would stop treatment and discharge the patient. Id. “Some . . . recruiters also ran halfway houses and

5

23-10182 Opinion of the Court 5

made extra money when HP referred discharged patients to those facilities.” Id.

“At trial, the recruiters explained that HP had them enter into contracts that stated they were providing either ‘case management ’ or ‘marketing’ services.” Kallen-Zury II, 710 F. App’x at 367– 68. “HP also asked the recruiters to submit reports documenting their purported performance of these services.” Id. at 368. “The recruiters’ reports, however, were false,” as the recruiters “were never asked to and never did provide these other services.” Id. Instead , the recruiters “were paid solely to refer patients.” Id. The government also produced various documents at trial, including a “patient register” that tracked which patients were referred by which recruiters. Id. at 370.

Several recruiters, each of whom testified under a grant of immunity, claimed to have discussed with Miller how HP would pay them to recruit patients and how HP would admit the recruited patients. For instance, Humes testified that when he had trouble admitting one of his recruited patients to HP, he called Miller , and she arranged for his patient to be admitted to HP. Himmons testified that, at first, she worked for Humes to recruit patients for HP, and Miller later asked her to send patients to HP through Veraguas. Eventually, Miller and Kallen-Zury hired Himmons to continue recruiting patients as a “marketer.” Miller also instructed Himmons to submit false monthly reports detailing her nonexistent marketing services. Moore testified that he met with Miller multiple times to get paid to recruit patients for HP and that

6

6 Opinion of the Court 20-11988

she instructed him to have his patients falsely claim that they were suicidal and off their medication to be admitted to HP. Miller also agreed to pay Moore through his non-profit company to make his payments appear legitimate. Veraguas similarly testified that he called Miller to get paid to recruit patients for HP.

The government also presented the testimony of Dr. Gumer to explain that most of the patients admitted to HP did not need psychiatric treatment, either because they only had substance abuse issues or because they were homeless. Kallen-Zury v. United States (Kallen-Zury IV), No. 20-12385, 2023 WL 164065, at *2 (11th Cir. Jan. 12, 2023). Dr. Marci Kagan and Marcia Starkman also testified for the government. Dr. Kagan, a psychotherapist at HP, testified that Miller told her to “stress the negative” in her patient files because, otherwise “the patient wouldn’t have criteria to be in the hospital.” Starkman, an HP employee, testified that she once learned patient files were being fabricated at HP and notified Miller about an incident where a doctor falsified a note and then tried to have the note shredded. Miller, however, never addressed the issue , and when Starkman tried to raise the issue in an administrative meeting, Miller jammed her elbow in Starkman’s side, signaling that Starkman should keep quiet.

After the government rested its case, the defendants, including Miller, argued that “they acted in good faith and believed the recruiters were providing lawful ‘marketing’ services.” Kallen-Zury I, 629 F. App’x at 897. They also argued that “HP’s lawyers drafted the contracts with the recruiters and instructed HP’s management

7

23-10182 Opinion of the Court 7

[on] how to ensure that their agreements with the recruiters fell within statutory and regulatory ‘safe harbor’ provisions.” Id. at 897–98.

Miller’s trial counsel, Malone, called Miller to testify in her own defense. Miller denied all the charges in the indictment. She testified that she never asked anyone to increase the severity of their notes to justify the patients’ continued stays at HP. She knew that HP paid some halfway-house owners for marketing services but did not believe that HP was paying these owners to send patients to HP. According to Miller, Veraguas and Humes “provided a lot of case management,” and Himmons “did business development ” for HP. Miller also testified that she did not determine how long patients stayed in HP’s inpatient facility and that HP’s psychiatrists decided when to admit and discharge patients. Miller disputed the testimony of all of the government’s witnesses regarding their interactions with her. Malone called no other witnesses.

As an important side note, the government did not receive patient files for the two patients listed in Counts 7 and 8. See Kallen- Zury IV, 2023 WL 164065, at *6 n.4. The government raised a Federal Rule of Criminal Procedure 16 objection to the use of the two patient files, which the defense expert, Dr. Jeffrey Danziger, had reviewed . Id. The district court sustained the objection and preemptively excluded any testimony on the two patients, such that Dr. Danziger could not testify on those two patients if he were called as a witness. Id. That said, Dr. Danziger was not called as a witness by Malone.

8

8 Opinion of the Court 20-11988

The jury found Miller guilty of all the charges. Kallen-Zury I, 629 F. App’x at 898. The judge sentenced Miller to fifteen years in prison. After her conviction, Miller filed a direct appeal, and we affirmed her conviction. See id.

B. Motion to Vacate

On March 25, 2016, Miller moved to vacate her convictions under 28 U.S.C. § 2255, asserting that Malone provided ineffective assistance under Strickland v. Washington, 466 U.S. 668 (1984). Of relevance to this appeal, Miller claimed that Malone failed to conduct an adequate investigation, consult and present an expert witness , and subpoena potential witnesses for trial. In support of her motion, Miller attached the following: affidavits from herself and her husband, Dale Miller; a twenty-nine-page summary of HP employees and their potential testimony; interview reports of prospective witnesses from Malone’s investigator; and many emails from Malone, Miller, and other individuals. Miller’s motion referenced a separate list of prospective trial witnesses that she provided Malone after he was appointed to represent her. Miller also filed affidavits from potential trial witnesses who were not called at trial.

The district court referred Miller’s motion to a magistrate judge. The magistrate judge then held an evidentiary hearing on Miller’s motion. The following testimony was elicited at the evidentiary hearing.

First, Robin Smith, HP’s utilization review coordinator and psychotherapist, stated that, if called as a witness, she would have testified that in her role as a coordinator, she reviewed patient files,

9

23-10182 Opinion of the Court 9

and she did not notice any inconsistencies among entries by psychiatrists , therapists, and others. According to Smith, Miller supervised the therapists but did not supervise HP’s psychiatrists, medical doctors, or nurses. Smith stated that patients were not admitted to HP if they did not have the appropriate psychiatric criteria for admission. She was unaware of any policy to keep patients at HP for the duration of their Medicare coverage, regardless of their psychiatric needs, or to discharge them when their Medicare coverage ended. On cross-examination, Smith conceded that she was not responsible for billing or marketing decisions. She did not know if HP’s “marketers” were recruiters who were paid to bring patients to HP. She was also not present for any meetings that Miller had with Moore, Humes, or Himmons.

Manuel Llano is a healthcare CEO who supervised Miller at Sunrise Regional Medical Center (“Sunrise”) from 1999 through 2001, before she started working at HP. Llano testified that he offered Miller a job at least twice after she left Sunrise, but she declined those offers because she enjoyed working at HP. No one contacted Llano before the trial about testifying on Miller’s behalf. On cross-examination, Llano conceded that he was not present at the meetings between Miller and HP’s recruiters, never saw Miller interact with patients at HP, never reviewed any of HP’s patient charts, and was unaware of any instructions HP gave its employees about completing patient charts.

Dr. Michael Piercey is the director of a medical facility that tests new medications and was the Chairman and CEO of Sunrise

10

10 Opinion of the Court 20-11988

when Miller worked there. Dr. Piercey testified that he was impressed by Miller’s intelligence, energy, and ethics, and would “be delighted to have [Miller] work with [him] again[,] . . . assuming she were available and interested.” Dr. Piercey was not contacted before the trial. On cross-examination, Piercey conceded that he was unfamiliar with HP’s admissions process, did not know whether HP paid recruiters to recruit patients, and did not know what Miller told Dr. Gumer and other HP employees.

Rita Sordellini worked as a part-time psychotherapist for HP. Whenever Sordellini reviewed patient files at HP, she never noticed inconsistencies among the notes entered by different individuals . She did not recall therapists raising concerns about inconsistencies at the weekly treatment team meetings that Miller facilitated . Sordellini never saw evidence of Miller instructing therapists to stress a patient’s negative symptoms to ensure that Medicare would reimburse HP. Miller never told her that particular patients needed to be discharged because they were out of Medicare days nor instructed her to keep patients longer because they had Medicare days remaining. Sordellini only had a “very quick” conversation with Malone or his investigator before trial and volunteered to testify for Miller, but she was not contacted afterward. On cross- examination, Sordellini conceded that Malone emailed her asking her to contact him about “anything or anyone that would be helpful to [Miller’s] cause.” Sordellini acknowledged that she was not responsible for discharge, insurance, or marketing decisions at HP and would not be aware if HP hired “marketers” to recruit patients.

11

23-10182 Opinion of the Court 11

She also conceded that she was not present at any meetings between Miller and the recruiters.

Melvin Hunter oversaw the Admissions Department at HP, and he testified that patients who arrived at HP were usually “precleared ” through an interview process and that doctors at HP ultimately made the final admission decision. He also testified that HP admitted patients even though there was no chance HP would be paid on “quite a few” occasions. As for his willingness to testify in Miller’s case, Hunter explained that he had declined to testify before the grand jury because his attorney was not permitted to accompany him into the grand jury room. Hunter, however, stated that he would have testified at Miller’s trial, even if his attorney advised him not to testify, because he believed that the allegations against HP and its staff were false. Hunter did not know if anyone contacted his attorney about testifying at Miller’s trial or if his counsel invoked his Fifth Amendment rights. On cross-examination , Hunter acknowledged that he did not know what happened to the patient files after the admissions process was completed or if they were falsified after he saw them. Hunter also conceded that he was not privy to all the conversations between Miller and Dr. Gumer, other staff members, or the recruiters.

Michael Calabria provided an affidavit but did not testify at Miller’s hearing. According to his affidavit, he worked with Miller as a psychotherapist at Sunrise and HP and presently works at HP’s successor facility. Calabria stated that Miller was a “great supervisor ” and a “great mental health professional.” Calabria also stated

USCA11 Case: 20-11988 Document: 84-1 Date Filed: 07/10/2023 Page: 12 of 30

12 Opinion of the Court 20-11988

that the quality of care at HP was “very good” and that the doctors were responsible for admitting and discharging patients. Miller never told him to write anything specific in his patients’ charts. Finally , he stated that no one contacted him about testifying at Miller ’s trial.

Sandra Novak also provided an affidavit but did not testify at Miller’s hearing. Novak worked with Miller as a therapist at Sunrise and HP. In Novak’s opinion, Miller would not have jeopardized her career, family, and future to participate in the charged offenses. No one contacted her about testifying at Miller’s trial.

Additionally, Roy Rindom provided an affidavit but did not testify at Miller’s hearing. Rindom worked with Miller as a psychotherapist at HP from 2004 to 2007. Rindom’s job was to conduct patient evaluations, including psychosocial interviews for admitted patients. Rindom stated that Miller never asked him to do anything unethical and that there were “no shenanigans” in connection with admissions, discharges, or treatment of patients. However, Rindom acknowledged that he was “engrossed in [his] daily duties and . . . was not aware of what Daisy Miller was doing at any given moment of the work day.” Rindom was also not involved in insurance -based decisions at HP. Finally, according to Rindom, no one contacted him about testifying at Miller’s trial. 1

1 Miller concedes, however, that contrary to Rindom’s affidavit, Malone interviewed Rindom before the trial.

13

23-10182 Opinion of the Court 13

Karen Bryan provided an affidavit but did not testify at Miller ’s hearing. She monitored patients, took vital signs, and assisted with the intake process. Bryan acknowledged that she had little interaction with Miller but thought Miller was “amazing.” However , she knew nothing about HP’s billing process. No one contacted her about testifying at Miller’s trial.

Miller also testified at the evidentiary hearing. After she was indicted, she prepared a list of witnesses she believed would help her defense and provided it to Malone. Miller was aware that Malone agreed to a joint defense agreement with her co-defendants , and they attended the weekly joint defense meetings together . But as the trial grew closer, Miller concluded that Malone’s loyalty was to the joint defense theme rather than to her individual defense. When Miller told Malone that she wanted a medical expert to testify on her behalf and suggested several psychiatrists, Malone told her that Kallen-Zury’s attorney, Michael Pasano, had retained an expert, Dr. Danziger, who would testify about the clinical issues. Malone never discussed with Miller the reason why Dr. Danziger did not testify.

On cross-examination, Miller conceded that she did not include Calabria, Hunter, Llano, Piercey, Sordellini, and Bryan as potential witnesses in the twenty-nine-page memorandum she prepared for Malone before trial. Miller also acknowledged that, contrary to Rindom’s affidavit, Malone had interviewed him before the trial. Miller also conceded that Malone contacted other witnesses on her behalf, sent her interview reports, and gave her periodic

14

14 Opinion of the Court 20-11988

updates. Finally, she conceded that none of her witnesses at the evidentiary hearing interacted with HP’s recruiters, with the possible exceptions of Humes and Veraguas.

Malone testified on behalf of the government. Malone is an experienced criminal defense attorney who has tried between fifty and sixty jury trials to verdict. Malone testified that, for this case, he conducted an extensive pretrial investigation, which entailed reviewing discovery at the government’s warehouse, examining patient files and other documents, hiring an investigator to obtain witness statements, meeting and sharing information with attorneys for Miller’s co-defendants, and identifying and interviewing potential witnesses. As for the joint defense agreement, Malone testified that he entered into the agreement in order to pool resources, jointly attack the government’s charges, identify who the government ’s witnesses might be, and take advantage of the work that the other defense attorneys had already done. As to his strategy for Miller’s defense, Malone testified that he intended to present Miller to the jury as “an upstanding citizen of this community who took her work very seriously” and “looked out for the well-being of [HP] patients.” He believed that Miller would be an excellent witness in this regard, and he decided not to call other witnesses so that the government would be unable to repeat, through cross-examination , the negative information it had presented in its case-in-chief. As to the allegations about Miller’s participation in recruiting patients , Malone testified that he did not have any evidence to refute them other than Miller’s general denial.

15

23-10182 Opinion of the Court 15

Next, as to his investigation of potential witnesses, Malone remembered speaking with Llano, could not recall who Dr. Piercey was, could not recall if he spoke with Smith, and recalled interviewing Sordellini but could not recall any details about their conversation . With respect to Hunter, Malone testified that he could not contact him directly because he was represented by counsel and that Hunter’s counsel would not let him testify at Miller’s trial. As for Dr. Danziger, Malone determined that Dr. Danziger could not have explained why so many out-of-state patients were admitted when they could have gotten the same services much closer to home. Malone noted that Dr. Danziger admitted that the out-ofstate patients were “a problem.” Malone was also concerned that Dr. Danziger would be cross-examined about how his opinion assumed that the patient files were not fabricated when Dr. Gumer had already testified that the patient files were fabricated.

The magistrate judge also took judicial notice of Dr. Danziger ’s testimony from the hearing on Kallen-Zury’s § 2255 motion . Kallen-Zury retained Dr. Danziger in May 2012 to review a sample of HP patient files. Over a one-year period, Dr. Danziger reviewed thirty-six charts of nineteen different patients to determine whether there were reasonable grounds for the patients’ admissions . At Kallen-Zury’s § 2255 hearing, Dr. Danziger testified that all nineteen patients met Medicare’s standards for admission to HP. As for suspicions of document fabrication, Dr. Danziger testified that it was unlikely that the charts he reviewed were fabricated because the entries were made by multiple practitioners and internally consistent, and “it would stretch credibility that

16

16 Opinion of the Court 20-11988

everyone was faking.” Still, Dr. Danziger conceded that he could not determine whether the notes he reviewed were truthful and could only determine that they were consistent across multiple medical practitioners. Finally, Dr. Danziger testified that he did not remember speaking with Malone. But contrary to his testimony at Kallen-Zury’s § 2255 hearing, Dr. Danziger’s declaration filed here states that on June 15, 2013, he spoke with Malone about his proposed trial testimony.

Finally, the magistrate judge took judicial notice of Pasano’s testimony from Kallen-Zury’s § 2255 hearing. Pasano’s trial strategy was to build a case around Kallen-Zury’s good faith in operating HP. Pasano testified that an important part of his defense strategy was to have Kallen-Zury testify and that he made a tactical decision not to call other witnesses once she completed her testimony . He had reservations about the limited nature of Dr. Danziger ’s testimony because Dr. Danziger only reviewed a few patient files. He also feared that, in cross-examining Dr. Danziger, the government could highlight its contention that the patient records had been falsified, rendering Dr. Danziger’s opinion incorrect for having assumed the veracity of the patient files.

On May 24, 2019, after the evidentiary hearing, the magistrate judge issued a 110-page report and recommendation, recommending that the district court deny Miller’s motion because Miller had not satisfied the performance prong or the prejudice prong under Strickland and also recommending that the district court issue

17

23-10182 Opinion of the Court 17

a certificate of appealability (“COA”) regarding Malone’s failure to call additional witnesses.

As for the performance prong of Strickland, the magistrate judge explained that Malone’s decision not to call additional witnesses , “made after balancing the potential risks against the possible benefits,” was “logical, rational, and understandable” and therefore “reasonable.” The magistrate judge noted that Malone had spent about 1,200 hours defending Miller and that, in reviewing counsel’s performance, a court must avoid using “the distorting effects of hindsight” and instead must evaluate the reasonableness of Malone’s performance from his perspective at the time. And the magistrate judge emphasized that Pasano had made exactly the same decision as Malone in not calling additional witnesses.

As to the prejudice prong of Strickland, the magistrate judge concluded that “there is no reasonable probability that [Miller’s] prosecution would have had a different outcome had Malone called” witnesses other than Miller at trial. Most of the witnesses could only offer impermissible character evidence about Miller, which the trial court had already excluded. Other witnesses were not “smoking gun” type witnesses. Several witnesses never worked at HP in the first place, so they could not testify about what Miller did or did not do at HP or what she did or did not know about the fraud at HP. Other witnesses did work at HP during the years at issue, but they did not know the inner workings of HP’s finance, marketing, insurance, and admissions departments. Still other witnesses were not at HP every day, were unaware of what Miller did

18

18 Opinion of the Court 20-11988

every day, or were not privy to the conversations that Miller had with those involved in the fraud.

On April 30, 2020, the district court adopted the “thorough, exhaustive, and persuasive” report and recommendation and denied Miller’s motion. The district court then issued a COA only as to Miller’s claim “that her former trial counsel was ineffective for failing to call any witnesses at trial other than [Miller] herself.”

This appeal ensued.

II. STANDARD OF REVIEW In a § 2255 proceeding, we review the district court’s legal conclusions de novo and the underlying factual findings for clear error . Lynn v. United States, 365 F.3d 1225, 1232 (11th Cir. 2004). “We allot ‘substantial deference to the factfinder . . . in reaching credibility determinations with respect to witness testimony.’” Devine v. United States, 520 F.3d 1286, 1287 (11th Cir. 2008) (quoting United States v. McPhee, 336 F.3d 1269, 1275 (11th Cir. 2003)).

III. ANALYSIS

A. Scope of Appeal Before reaching the merits of this case, we must address a procedural issue. The report and recommendation, which the district court adopted, states that “[t]rial counsel’s decisions, particularly about whether to call certain witnesses and introduce documents , are entitled to great deference, and there is a presumption that trial counsel’s decisions are reasonable.” However, Miller argues that Malone’s decisions should not be entitled to deference

19

23-10182 Opinion of the Court 19

because they were based on an inadequate investigation and were thus not informed, strategic decisions. Thus, according to Miller, our inquiry must include an analysis of whether Malone’s investigation “was itself reasonable.” That is, even though the COA speci fies that the only issue on appeal is whether Malone “was ineffective for failing to call any witnesses at trial other than [Miller] herself ,” Miller argues that the Court should address a subsidiary claim that Malone failed to conduct an adequate investigation of potential witnesses and documents. The government agrees that addressing the issue on appeal necessarily encompasses the subsidiary claim of whether Malone conducted an adequate investigation of other potential witnesses. However, to the extent that Miller argues that Malone failed to properly review the government’s discovery and investigate potential documents—rather than potential witnesses—the government contends that such arguments are beyond the scope of the COA and that we should not address those issues.

Although appellate review is limited to issues specified in the COA, Turner v. Sec’y, Dep’t of Corr., 991 F.3d 1208, 1211 n.1 (11th Cir. 2021) (quoting Murray v. United States, 145 F.3d 1249, 1250–51 (11th Cir. 1998)), we conclude that Malone’s decision to call witnesses could have been elucidated by a reasonable investigation of the government’s discovery and documentary evidence. Thus, our inquiry will include an analysis of whether Malone’s investigation of the documentary evidence, as well the potential witnesses, was reasonable to determine whether Malone’s decisions at issue are entitled to deference.

20

20 Opinion of the Court 20-11988

With this procedural matter resolved, we now turn to the merits of Miller’s ineffective assistance claim.

B. Prejudice

To prevail on an ineffective assistance of counsel claim, a petitioner must “demonstrate both that (1) ‘counsel’s performance was deficient,’ and (2) ‘the deficient performance prejudiced the defense .’” United States v. Webb, 655 F.3d 1238, 1258 (11th Cir. 2011) (quoting Strickland, 466 U.S. at 687). “We may consider the prongs of the Strickland test in either order, and the defendant must show that both prongs are satisfied in order to demonstrate a Sixth Amendment violation.” Id. This standard is necessarily a fact-intensive one, requiring a court to “consider[] all the circumstances” when evaluating counsel’s performance. Strickland, 466 U.S. at 688. If a defendant makes an insufficient showing on one of the requisite prongs, we need not address the other prong. Id. at 697. Determining whether there is prejudice under Strickland is a mixed question of law and fact, making our review plenary. See, e.g., Brooks v. Comm’r, Ala. Dep’t of Corr., 719 F.3d 1292, 1300 (11th Cir. 2013). Underlying factual determinations, however, are subject to clear error review. See Devine, 520 F.3d at 1287.

To establish prejudice under the second prong of Strickland, a defendant must show “a reasonable probability that, but for counsel ’s unprofessional errors, the result of the proceeding would have been different.” Id. at 694. “A reasonable probability is a probability sufficient to undermine confidence in the outcome.” Id. In other words, “[i]t is not enough for the defendant to show that the

21

23-10182 Opinion of the Court 21

errors had some conceivable effect on the outcome of the proceeding .” Id. at 693. The petitioner instead must show that counsel’s deficient representation rendered the result of the trial unfair. See id. at 697. The prejudice component of the Strickland standard thus reflects “[t]he purpose of the Sixth Amendment guarantee of counsel ,” which is to “ensure that a defendant has the assistance necessary to justify reliance on the outcome of the proceeding.” Id. at 691–92. We have recognized that, given the principles and presumptions associated with ineffective assistance claims, “the cases in which habeas petitioners can properly prevail are few and far between.” Chandler v. United States, 218 F.3d 1305, 1313 (11th Cir. 2000) (en banc) (omission adopted) (quoting Waters v. Thomas, 46 F.3d 1506, 1511 (11th Cir. 1995) (en banc)).

Here, we conclude that the district court did not err in determining Miller did not suffer prejudice by Malone’s decision not to call additional witnesses. The magistrate judge found that Miller ’s proposed witnesses were either not credible or credible but could not offer material testimony, meaning their testimony would not have changed the outcome of the trial. After a review of the record, we conclude that the magistrate judge’s factual findings about the potential witnesses are not clearly erroneous, especially given the substantial deference we afford to the factfinder as to credibility determinations. We discuss each proposed witness in turn.

First, Dr. Danziger’s testimony would have been of little value to Miller’s defense. As we noted previously in Kallen-Zury IV,

22

22 Opinion of the Court 20-11988

because Dr. Danziger analyzed only nineteen of the thousands of patients at HP during the period in question, his conclusions were not based on a significant sample to hold much weight at trial. See 2023 WL 164065, at *9. “Additionally, his analysis assumed the veracity of the treatment forms, even though he had no corroboration for such an assumption.” Id. Dr. Danziger conceded that, while the files were internally consistent, he could not determine whether the files were truthful, raising the possibility that the files were consistently falsified. Even if the jury believed Dr. Danziger’s claim that it was unlikely that everyone had falsified their part of the patient files in an internally consistent manner, he could not explain the out-of-state patients, which he described as a “problem .” In addition, Dr. Danziger could not rebut the recruiters’ testimony and documentary evidence about Miller’s participation in hiring recruiters and preparing the recruiters’ false monthly reports . So even if he could convince the jury that the recruited patients were properly admitted, he would not have been able to refute the evidence that HP was not permitted to pay recruiters to bring patients but did so anyway. Being unable to address that aspect of the criminal scheme and Miller’s participation in that scheme, there is no reasonable probability that Dr. Danziger’s testimony would have led to a different outcome here.

Next, Smith was not responsible for billing and marketing and did not know if the recruiters were paid to recruit patients. She testified that she did not attend any meetings that Miller had with Moore, Humes, or Himmons. Therefore, like Dr. Danziger, she would have not refuted Miller’s involvement in the recruiting

23

23-10182 Opinion of the Court 23

scheme. She would have merely testified that she did not know of any fraud at HP. Her lack of knowledge, however, would not have overturned the verdict. Indeed, it is unsurprising that those participating in a fraudulent scheme would not inform all of their employees of the fraud. Thus, the jury could have reasonably concluded that Miller and her co-defendants took affirmative steps to keep knowledge of the fraud from Smith and other non-participating employees.

Similarly, Sordellini’s potential testimony that the patients were properly admitted would not have overturned the verdict. As a part-time employee at HP, Sordellini conceded that she was not responsible for discharge, insurance, or marketing decisions at HP, and she was unaware if HP hired “marketers” to recruit patients. And Sordellini conceded that she was not present at any meetings with Miller and the recruiters. It is therefore unlikely that the jury would have rejected the government’s evidence of fraud and Miller ’s participation in the fraud based on a part-time employee’s testimony that she was unaware of the scheme. Instead, it is more likely that the jury would have concluded that, as a part-time employee , Sordellini was simply not privy to the criminal scheme. Thus, like Dr. Danziger and Smith, she would not have refuted the government’s evidence of Miller’s involvement.

As for Hunter, we already concluded in Kallen-Zury IV, that his testimony would not have impacted the outcome of the trial with respect to Kallen-Zury. See 2023 WL 164065, at *8. The same conclusion applies to the outcome of the trial with respect to

24

24 Opinion of the Court 20-11988

Miller. Hunter conceded that he was unaware if Miller instructed Dr. Gumer to falsify records to ensure that patient files reflected severe symptoms. Similar to the previous witnesses, the fact that he was unaware does not mean that the jury would have rejected the government witnesses’ testimonies. Being unaware of the fraud does not establish that the alleged conduct did not happen; it merely suggests that Hunter did not know whether the conduct happened. Furthermore, similar to the magistrate judge in Kallen- Jury IV, 2023 WL 164065, at *8, the magistrate judge here did not find Hunter’s testimony to be credible. We agree that Hunter’s claim that he would have testified at Miller’s trial is not credible because he invoked his Fifth Amendment rights before the grand jury. In sum, Hunter lacked credibility and was not privy to the criminal scheme to have made a difference in Miller’s trial.

As for Llano and Dr. Piercey, because they never worked at HP, they too could not credibly testify about the fraud at HP. To the extent that they could have undercut the government’s theory that Miller engaged in criminal conduct to keep her job by testifying that she had other job prospects, their testimony about her motive , or the lack thereof, would not have overwhelmed other documentary and testimonial evidence of her criminal conduct. The government presented significant evidence over the five-week trial, and Llano and Dr. Piercey’s ancillary arguments related to motive would have been insignificant in comparison. Thus, even if their testimonies were admitted, it is unlikely that they would have made a difference.

25

23-10182 Opinion of the Court 25

The affiants, Calabria, Novak, Rindom, and Bryan, would not have made a difference either. Even if the affiants testified that they were unaware of the criminal scheme, that does not exclude the possibility that Miller asked other employees to participate in the scheme without the affiants’ knowledge. In addition, Novak and Bryan would have merely provided character evidence that Miller was honest, which would not have overturned the outcome of the trial.

In sum, the testimony from these witnesses—taken together or separately—would not have created a reasonable probability that Miller’s trial would have resulted in a different outcome. See Devine, 520 F.3d at 1287. None of the potential witnesses could rebut the government’s evidence that Miller participated in a kickback scheme with the recruiters because they were not privy to the pertinent conversations between Miller and the recruiters. Thus, and considering the weight of the evidence presented against Miller during the five-week trial, we conclude that Miller cannot show that she was prejudiced by Malone’s failure to call these witnesses at trial and cannot satisfy the second Strickland prong.

C. Deficient Performance Because Miller has failed to show that she was prejudiced by counsel’s failure to call witnesses at trial, we need not consider the first prong of Strickland. However, even if we were to assume that Miller were prejudiced and met the second Strickland prong, Miller still cannot prevail because Malone’s decision not to call additional witnesses did not amount to deficient performance.

26

26 Opinion of the Court 20-11988

“The standard for counsel’s performance is ‘reasonableness under prevailing professional norms.’” Chandler, 218 F.3d at 1313 (quoting Strickland, 466 U.S. at 688). There is a strong presumption that counsel rendered reasonable professional assistance and thus Miller has the burden to overcome this presumption. See Dingle v. Sec’y for Dep’t of Corr., 480 F.3d 1092, 1099 (11th Cir. 2007). We generally afford great deference to counsel’s strategic decision not to call witnesses. Conklin v. Schofield, 366 F.3d 1191, 1204 (11th Cir. 2004); Chandler, 218 F.3d at 1313. Further, “[o]ur strong reluctance to second guess strategic decisions is even greater where those decisions were made by experienced criminal defense counsel.” Provenzano v. Singletary, 148 F.3d 1327, 1332 (11th Cir. 1998). And “[e]ven if counsel’s decision appears to have been unwise in retrospect , the decision will be held to have been ineffective assistance only if it was ‘so patently unreasonable that no competent attorney would have chosen it.’” Dingle, 480 F.3d at 1099 (quoting Adams v. Wainwright, 709 F.2d 1443, 1445 (11th Cir. 1983)).

As noted above, we must first determine whether Malone’s investigation was adequate before giving great deference to his decision not to call witnesses. See Wiggins, 539 U.S. at 521–23. Turning to Malone’s investigative efforts, the magistrate judge found that Malone, an experienced criminal defense attorney, spent about 1,200 hours defending Miller.

Although Malone did not speak to Dr. Danziger until after Miller’s trial began, Malone testified that he knew what Dr. Danziger would have discussed at trial beyond “best practices,”

27

23-10182 Opinion of the Court 27

indicating that he was sufficiently apprised of Dr. Danziger’s testimony through the work of the joint defense team. Malone’s decision not to have another medical expert review additional files re- flected Malone’s reasonable investigation of the documents and Malone’s understanding that a second medical expert would not have been able to rebut any allegations about Miller’s supervision of the recruiters from the patient files, thereby making any additional investigation into patient files unnecessary and unproductive in defending Miller. And as for the remaining witnesses who testi- fied that Malone did not contact them or only made a brief call, we have explained that counsel “need not interview every conceivable witness” to satisfy Strickland. Morrow v. Warden, 886 F.3d 1138, 1148 (11th Cir. 2018); see also Chandler, 218 F.3d at 1317 (“[N]o absolute duty exists to investigate particular facts or a certain line of defense .”). Malone realized that none of those witnesses could rebut the critical evidence about the recruiters based on the detailed list of witnesses that Miller provided Malone. Finally, with regard to Hunter, he was represented by counsel, and Malone cannot be faulted for not contacting him directly or his counsel given that Malone does not have to pursue all leads, especially not a witness who had previously invoked his Fifth Amendment rights. Cf. Davis v. Laffler, 658 F.3d 525, 537 (6th Cir. 2011) (en banc) (concluding that counsel’s decision not to call a witness was a strategic choice based in part on the concern that witness “would exercise his Fifth Amendment right to remain silent if called”).

Even if it is otherwise unclear how Malone spent the 1,200 hours, whether he effectively used that time reviewing documents

28

28 Opinion of the Court 20-11988

and interviewing additional witnesses, and why he did not discover that the government did not have certain patient files during that time, Malone’s inability to recollect the specifics of his investigation six-and-a-half years after the fact is not a sufficient ground to conclude that his investigation was inadequate. Blankenship v. Hall, 542 F.3d 1253, 1274 (11th Cir. 2008) (“[W]hen the evidence is unclear or counsel cannot recall specifics about his actions due to the passage of time and faded memory, we presume counsel performed reasonably and exercised reasonable professional judgment.”); Harvey v. Warden, Union Corr. Inst., 629 F.3d 1228, 1245 (11th Cir. 2011) (similar). As we stated previously, attorneys are entitled to the presumption that their conduct was reasonable, and “[a]n ambiguous or silent record is not sufficient to disprove the strong and continuing presumption.” Chandler, 218 F.3d at 1314 n.15.

We thus conclude that Malone sufficiently investigated the witnesses and documents. Thus, his decision not to call additional witnesses was a strategic choice made after an adequate investigation , which makes his decision “virtually unchallengeable.” Strickland , 466 U.S. at 690; accord Conklin, 366 F.3d at 1204 (“Which witnesses , if any, to call, and when to call them, is the epitome of a strategic decision, and it is one that we will seldom, if ever, second guess.” (quoting Waters, 46 F.3d at 1511)).

With this deferential standard in mind, we turn to whether Malone’s explanations for not calling witnesses were reasonable. Malone, an experienced criminal defense attorney, testified that he did not call Dr. Danziger because Dr. Danziger would have faced

29

23-10182 Opinion of the Court 29

damaging cross-examination. Dr. Danziger reviewed only a few patient files and assumed the veracity of the patient files, which would have been revealed through cross-examination. And, as stated before, the government’s theory was not that Miller was only involved in the patient side of HP. The government’s theory was that Miller directed recruiters, which goes beyond patient care, and Dr. Danziger would not have been able to credibly testify about this allegation based on his limited review of patient files. In addition , Dr. Danziger would not have been able to provide a satisfactory explanation for out-of-state patients coming to HP, but for the work of Miller’s recruiters. Thus, Malone’s decision not to call Dr. Danziger was reasonable.

As for Malone’s decision not to call other witnesses, all of the potential witnesses would have similarly been unable to refute critical aspects of the government’s case and, as a result, would have been subject to damaging cross-examination. See Chandler, 218 F.3d at 1321 (“[A] lawyer reasonably could also fear that character evidence might, in fact, be counterproductive: it might provoke harmful cross-examination and rebuttal witnesses. Misgivings about hurtful cross-examination and rebuttal witnesses have been decisive to the Supreme Court when it determined that counsel was effective.” (footnote omitted)). Thus, we conclude that Malone’s decision not to call additional witnesses was reasonable, and that Miller has not adequately demonstrated that Malone was deficient in his representation.

30

30 Opinion of the Court 20-11988

IV. CONCLUSION

For all these reasons, we affirm the district court’s denial of Miller’s § 2255 motion to vacate her conviction.

AFFIRMED.