Ralph F. Vitale appeals from his conviction and sentence entered after a jury found him guilty of bank fraud in violation of 18 U.S.C. §§ 1344(1) and (2). Vitale participated in a scheme with Charles Hoblin and Peter Trantino to submit fraudulent business loan applications to Fleet Bank. Vitale principally argues that: (i) the district court violated his Sixth Amendment right to confront and cross-examine a witness by denying him access to the witness’ substance abuse treatment records; (ii) the district court erred in refusing to conduct a post-trial inquiry into juror bias after revelation of a professional relationship betwеen the juror, the juror’s husband, and the prosecutor’s husband; and (iii)
United States v. Fagans,
We conclude that the district court committed no Sixth Amendment violation by denying access to the government witness’ substance abuse treatment records. However, we remand for a hearing on the issue of possible juror bias under the procedure set out in
United States v. Jacobson,
BACKGROUND
We view the evidence in the light most favorable to the government.
Glasser v. United States,
From about January 1996 to about March 1996, Vitale, Hoblin, and Trantino pаrticipated in a scheme to defraud Fleet Bank. They took advantage of Fleet Bank’s Easy Business Banking Loans— which provided to small businesses loans *192 of up to $100,000 through an expedited process with limited documentary requirements — by submitting fraudulent loan applications. Hoblin was a self-employed accountant who prepared individual income tax returns for his clients. He and Vitale concocted the fraudulent loan documents by creating phony corporations with fictitious financial information 1 and using names, addresses, tax identification numbers, and income tax returns of Hoblin’s unknowing clients to fill in the gaps. Tran-tino, who was a business development/loan officer at Fleet Bank, would sign each application as a witness and forward the application to the underwriting unit, knowing that the information being submitted was false and/or obtained without Hoblin’s clients’ permission. Working together, Vi-tale, Hoblin, and Trantino secured five Easy Business Banking Loans, each for the maximum $100,000.
As each loan was nearing approval, a bank account was opened for each entity. Two persons were designated to issue checks on each account: on four of the accounts Vitale (or a relative of Vitale’s) and the person in whose name the loan application had been submitted were authorized to draw on the funds; on the fifth account, a relative of Hoblin’s was named along with the unwitting “applicant.” Once the monies were deposited in the accounts, Vitale, Hoblin, and Trantino made withdrawals totaling $403,000.
On May 11, 2004, a jury found Vitale guilty of all five counts of bank fraud, and on August 16, 2004, Vitale was sentenced by the district court.
a) Trantino’s Substance Abuse Treatment Records
Trantino, who pled guilty to charges of bank fraud was a government witness. Prior to trial, the government informed defense counsel that Trantino had abused narcotic pain killers during the relevant period of time and that he currently was enrolled in a drug abuse treatment program.
Defense counsel subpoenaed Trantino’s records from the Orchard Drug Rehabilitation Clinic. The APT Foundation, which owned and operated the Orchard Drug Rehabilitation Clinic, moved to quash the subpoena, arguing that the information was protected from disclosure.
Vitale argued that because Trantino had admitted to the heavy use of painkillers during the bank fraud scheme and drug addiction when he was discussing the case with thе FBI, the substance abuse treatment records were relevant to Trantino’s ability to recall and relate pertinent events as well as to his general credibility. He argued that access to the records was necessary to cross-examine Trantino. Vitale’s counsel also speculated that Trantino’s referral to a substance abuse treatment program after his then-recent drug-related arrest was facilitated by the government, raising a concern of possible bias. The government assured the court that it had not intervened in Trantino’s state drug charges. The district court then conducted an in camera review of the records and reserved its rulings.
During his direct examination, Trantino testified that he had sustained a neck injury in mid-summer 1995. He was prescribed and took vicodin for this injury until early 1996, when he had surgery to repair his neck. After the surgery, Tran-tino was prescribed another narcotic pain *193 killer, Tylox, which he used for another six-to-eight weeks. Trantino claimed that neither painkiller affected his performance at the bank.
After this portion of the direct examination, the district court, over the objection of Trantino’s counsel and after its own in camera review, summarized for the parties the information contained in Trantino’s substance abuse treatment records. The court stated that it wanted Vitale’s counsel to have all “arguably germane” information in order for the defense to “fully and thoroughly cross-examine [Trantino] consistent with the right to confrontation.”
Outside the presence of the jury, the district court related that “starting at age 40 at some point in 1998, long after the transactions at issue here, Mr. Trantino began to use heroin.” The court then summarized Trantino’s lengthy substance abuse treatment: (i) in mid- to late-1998, outpatient detoxification treatment for four days, which occurred around the time of his initial contact with the FBI; (ii) inpatient treatment for three days in early 2000; and (iii) further inpatient treatment in mid-2001. The records indicated that Trantino last used heroin in March of 2002. Since that time, Trantino had received methadone treatments, although it was unclear whether those continued at the time of trial. The district court further revealed that Trantino had abused cocaine occasionally from the time he was twenty-five years old until his 2001 treatment for heroin but that the records were not clear as to the quantity or frequency of cocaine usage. Summing up, the distriсt court stated:
I think that’s a reasonably accurate summary of a rather voluminous record. And I will leave it to you to proceed from there.
If Mr. Trantino testifies consistently with that information, then I would see no reason why I should order that the documents be disclosed to you. If that doesn’t happen, then maybe there would be good cause or adequate cause for some further disclosure.
Defense counsel then moved that the physical records be disclosed pursuant to Vi-tale’s rights under the confrontation clause and that the records be sealed and marked for appellate review. The district court did not issue a final ruling on the matter, preferring to “see how it unfolds.”
Trantino admitted to the information summarized by the judge, and the district court made no additional disclosures. During cross-examination, defense counsel again questioned Trantino about his drug use and rehabilitation.
b) Post-Trial Juror Contact
On April 13, 2004, jury selection for another, unrelated case was being conducted while the pool of jurors for Vitale’s case, including juror Barbara Setlow, was probably present in the courtroom. During voir dire for the first case a potential juror remarked that she thought Assistant United States Attorney Lisa Perkins was related to someone she knew who had a different last name. Perkins stated that her husband’s name was Sparkowski. The juror indicated that she knew Perkins’ in-laws. Another potential juror raised his hand and asked Perkins if her husband’s name was Jason Sparkowski, indicating that he had attended high school with a Jason Sparkowski. Perkins revealed that her husband was indeed Jason Sparkow-ski. Jason Sparkowski was not discussed at any other time during voir dire for the first case.
During the voir dire in Vitale’s ease, Jason Sparkowski’s name was never mentioned. When Setlow introduced hеrself to the court, she stated that she was “a biochemist with the UConn Health Center” *194 and that her “husband works at the same place and does the same thing.” Later events, described below, revealed that Jason Sparkowski, Perkins’ husband, was also a biologist at the University of Connecticut Health Center (“UCHC”). Nevertheless, Perkins remained silent as Set-low was selected as a juror.
On May 17, 2004, following the jury verdict, Perkins sent a letter to Judge Chatigny disclosing the following. Jason Sparkowski had been a student in the biochemistry graduate program at UCHC from 1984-1990. During this time, Dr. Peter Setlow, the juror’s husband, was a professor in the biochemistry department. While Sparkowski was a student at UCHC, he had frequent contact with Dr. Setlow but never socialized with him. He did, however, know Barbara Setlow to be the spouse of Dr. Setlow.
According to the letter, Sparkowski left Connecticut in 1990 and lived in Washington, D.C. until 2000. During these ten years, Sparkowski had no contact with the Setlows. Sparkowski and Perkins married in 2000. In June 2003, Sparkowski returned to UCHC as an assistant research professor but only had brief, passing conversations with Dr. Setlow.
The letter stated that, on May 7, 2004, Sparkowski attended Vitale’s trial for approximately forty minutes. He thought he recognized Setlow, the juror, as an acquaintance of Dr. Setlow. Perkins did not know that Sparkowski attended the trial until that evening. Whether he mentioned his recognition of the juror was not revealed in the letter. After the conclusion of the trial, Perkins called Sparkowski to inform him of the jury’s verdict. This prompted Sparkowski to go to Dr. Setlow’s lab to ask him if his wife had recently served on a jury. When Sparkowski arrived at Dr. Setlow’s lab, both Setlows were there. Barbara confirmed that she had recently served on the jury, and Spar-kowski indicated that he had recognized her when he observed the trial. According to Perkins’ letter, Juror Setlow made no mention of having seen or recognized him, and no details of the jury deliberation were revealed in this conversation.
Based on Perkins’ letter, the defense moved for an evidentiary hearing on potential juror bias. Vitale’s counsel argued that Perkins’ letter was a “one-sided view of what occurred” — no affidavits from Jason Sparkowski or either of the Setlows were submitted to the court — and that it was appropriate for the court to further investigate the matter. Defense counsel made it clear that he was not “аccusing anyone of anything at this point” and that no motion for a new trial had been made. He simply wanted to call Sparkowski and the Setlows to obtain their version of the events and the relationships between them. Defense counsel also indicated that had he been aware of the professional relationship between Sparkowski and the Setlows that he would have moved to disqualify Setlow for cause.
After also hearing argument from Perkins, the court issued the following decision:
As a practical matter, a judge in this situation might think that the better part of valor would be to simply go ahead and call the juror in and see what she has to say. I’m tempted to do that because I have little doubt that we could resolve this very quickly.
On the other hand, the Court of Appeals has rightly emphasized that such post trial inquiries of jurors should not be undertaken lightly. I agree that there needs to be a substantial showing to warrant that type of inquiry no matter how convenient it might be to the Court.
*195 And applying the standard articulated in the Moon case [United States v. Sun Myung Moon,718 F.2d 1210 , 1234 (2d Cir.1983)], I think the defendant’s request should be denied. As [defense counsel] properly acknowledges, he isn’t saying that there was an impropriety, he doesn’t know that there was. He asks me to cоnduct this inquiry to find out if maybe there was some kind of an impropriety. It is entirely speculative. And I do think it is implausible to support that there was an impropriety.
In that context, I think this coincidental post trial contact is just that, and I don’t think it warrants a postponement of the sentencing hearing.
So the request for an evidentiary hearing to determine the existence of juror bias is denied....
DISCUSSION
Vitale advances three arguments on appeal. He contends that the district court’s denial of access to the substance abuse treatment records violated his Sixth Amendment rights to confront аnd cross-examine Trantino, that the district court erred by refusing to hold an evidentiary hearing on juror bias, and that he is entitled to a resentencing. We address each issue in turn.
a) Vitale’s Sixth Amendment Right to Confrontation and Cross-Examination
Alleged violations of the Confrontation Clause are reviewed
de novo,
subject to harmless error analysis.
United States v. Saget,
“The Sixth Amendment to the Constitution guarantees the right of an accused in a criminal prosecution ‘to be confronted with the witnesses against him.’ ”
Davis v. Alaska,
Restricting access to medical or psychological records may deny a defendant the right to confrontation because “[e]vidence of a witness’s psychological history may be admissible when it goes to her credibility.”
Sasso,
However, “the failure to produce a psychiatric report does not in itself effect a constitutional deprivation where ... the report does not contain evidence of any deep or sustained mental problems which would directly bear upon the credibility of the witness.”
White v. Jones,
We see no error in the denial of access to Trantino’s substance abuse treatment records. Although defense counsel was not allowed to review the physical records, “the trial court did not limit the scope or nature of defense counsel’s cross-examination in any way.”
Delaware v. Fensterer,
The court permitted Vitale’s counsel wide latitude when cross-examining Tran-tino about his drug use and rehabilitation, including questions about the effects that the drugs had on his ability to perceive events when they occurred as well as on his memory at the time of trial. The leeway granted was generous given that the records reflected Trantino’s treatment severаl years after the pertinent events and had marginal probative value concerning Trantino’s mental capacity during the events about which he testified.
Cf. Sasso,
Physical access to all the drug rehabilitation treatment records would not have provided defense counsel additional avenues to impeach Trantino’s credibility. Vi-tale was therefore not denied the right “to expose to the jury the facts from which jurors ... could appropriately draw inferences relating to the reliability of the witness,”
Davis,
*197 b) Juror Bias
The Supreme Court “has long held that the remedy for allegations of juror partiality is a hearing in which the defendant has the opportunity to prove actual bias.”
Smith v. Phillips,
We have also stated, however, that district courts should be reluctant “to haul jurors in after they have reached a verdict in order to probe for potential instances of bias, misconduct or extraneous influences,”
United States v. Sun Myung Moon,
Nevertheless, a trial court is required to hold a post-trial jury hearing when reasonable grounds for investigation exist.
Moon,
We believe that, on these facts, the district court was right in its initial belief that the “better рart of valor” would have been to hold an evidentiary hearing. While it is likely that nothing would have come of such a hearing, the lack of one has left too many unanswered questions and too much room for more surprises to occur. In particular, there are questions about why this issue was not aired during *198 the voir dire and whether the prosecutor knew during the trial that her husband recognized a juror. We also believe that requiring a hearing may create incentives for the government to address such matters before rather than after a verdict.
If the relationship between Sparkowski and the Setlows while Sparkowski was a student was at issue, we would of course find that no abuse of discretion occurred.
See, e.g., Phillips,
We know that Sparkowski was sufficiently interested in the case to attend the trial for a period of time, recognized the juror, and told his prosecutor wife that he had attended the trial. We know that he further sought out the juror’s husband after receiving his wife’s call about the verdict. He encountered the juror in her husband’s office or lab. It is said that no discussion of the jury’s deliberations occurred. It is also said that the juror, when encountering Sparkowski, gave no indication that she recognized Sparkowski when he was at the trial, an omission that is apparently the basis for the prosecutor’s letter’s later flat assertion that the juror “did not recognize” Sparkowski at the trial.
However, there are many unanswered questions. First, much of what we know about Sparkowski’s role is by way of hearsay. The prosecutor’s letter is based on no firsthand knowledge of her husband’s encounter with the juror, and we have no firsthand account from him of the encounter or of his motivation in seeking out the juror’s husband. We have no evidence from either Sparkowski or his prosecutor wife of their convеrsations in the evening of the day on which he attended the trial and whether he told her that he had recognized a juror. We also have no firsthand account from the juror of relevant events, only inferences to be drawn from the prosecutor’s recitation of her husband’s recitation of the meeting.
Further questions arise from the events of which we do have firsthand knowledge, namely the jury selection. The juror plainly stated that she and her husband were biochemists at UCHC. The prosecutor’s husband was also a biologist at UCHC, but she said nothing. The common workplace alone greatly increased the chance of a pre-existing or existing relationship between the three scientists. Even a juror who has a workplace friendship with a relative of an attorney may believe that revelation of her employment is sufficient to alert the parties. Moreover, the *199 attendance of the prosecutor’s husband at the trial greatly increased the chance of a subsequent encounter between the prosecutor’s husband and the juror.
If it was appropriate to report such an encounter to the court when it happenеd— and we believe it was — it was even more appropriate to inform the court at jury selection that the prosecutor’s husband was also a biologist at UCHC. Had the court and the defense been made aware of the fact that Juror Setlow, Dr. Setlow, and Sparkowski were all working at UCHC, defense counsel could have moved to strike the juror for cause, and the court could have granted the motion or questioned the juror about her ability to remain impartial despite the connections. In any event, the problem could have been addressed in a timely fаshion.
Of course, this is all hindsight — the source of much appellate court wisdom— but, while we realize that the problem here is likely the result of a confluence of innocent events, the appearance created by the present record does not exclude less-innocent inferences drawn from the prosecutor’s failure to disclose Jason Sparkowski’s employment as a biologist at UCHC during voir dire, or, if known, from Sparkow-ski’s recognition of a juror during trial. Were one more new and surprising fact to emerge in the future, more serious allegations could arisе that might be difficult to resolve with the decline of memory or the unavailability of key participants. If so, the hindsight available then might make our failure to order a hearing now seem like willful blindness.
In
Phillips,
the Supreme Court held that a prosecutor’s failure to disclose until after trial that a juror had applied to be an investigator in the district attorney’s office “requir[ed] a post-trial hearing on juror bias.”
Given these authorities, we believe that the totality of the circumstances — including the juror’s statement regarding her and her husband’s employment during the voir dire, Perkins’ failure to disclose Spar-kowski’s employment, Sparkowski’s attendance at trial and recognition of Setlow, Sparkowski’s contact with the Setlows after trial, and the various unknowns that exist — require a further investigation.
3
*200
We therefore remand this matter so that a post-trial evidentiary hearing may be held. We note, however, that we leave it to the discretion of the district court to determine the scope of the hearing.
See Ianniello,
We use the procedure set out in
Jacobson,
c) Sentencing
Vitale argues, and the government concedes, that Vitale’s sentence, imposed under the mandatory Sentencing Guidelines, violated
United States v. Booker,
CONCLUSION
We conclude that the district court committed no Sixth Amendment violation by denying physical access to Trantino’s substance abuse treatment records. However, we remand for an evidentiary hearing on juror bias. Wе follow the procedure set out in
Jacobson.
A mandate shall be issued forthwith restoring jurisdiction to the district court to hold an evidentiary hearing consistent with this opinion. After the district court’s decision, jurisdiction may be restored to this court by letter from any party, and the Clerk’s Office shall set a briefing schedule and send such proceeding to this panel for disposition without oral argument unless otherwise ordered. Further, pursuant to
Fagans,
we vacate Vitale’s sentence and instruct the district court to resentence him.
Notes
. Only three of the five loans were obtained for completely fictitious corporations. Two corporations used in the applications existed on paper, but the information supplied on behalf of these corporations was falsified.
. These cases relied on by Vitale are inappo-site. In
United States v. Lindstrom,
. District courts have found reasonable grounds to hold a post-trial evidentiary hearing on juror bias in circumstancеs with much more tenuous evidence of juror bias or misconduct than is present here.
See e.g., United States v. Shaoul,
