United States v. Ignasiak, Jr., No. 09-10596 (11th Cir. 2012)

Annotate this Case
Justia Opinion Summary

Defendant, a medical doctor licensed by the State of Florida, appealed his convictions for dispensing controlled substances in violation of the Controlled Substances Act (CSA), 21 U.S.C. 801, et seq., and for health care fraud. Defendant's convictions resulted in three separate but related appeals including his contention that the district court erred or abused its discretion by allowing the introduction of autopsy reports or handwritten medical notes without requiring testimony by their authors. The court reversed defendant's convictions because the admission of the autopsy reports and testimony about those reports, without live in-court testimony from the medical examiners who actually performed the autopsies, violated the Confrontation Clause under the facts of the case. Because the court concluded that this issue was dispositive, the court declined to address the other issues raised in defendant's merits appeal, except for the sufficiency of the evidence claim. While the court ultimately concluded that the evidence was sufficient, the degree to which the court viewed the government's case as less than overwhelming compelled the court's conclusion that the Confrontation Clause violation was not harmless in this case.

Download PDF
[PUBLISH] IN THE UNITED STATES COURT OF APPEALS FOR THE ELEVENTH CIRCUIT FILED ________________________ U.S. COURT OF APPEALS ELEVENTH CIRCUIT JANUARY 19, 2012 JOHN LEY CLERK Nos. 09-10596, 09-16005 & 10-11074 ________________________ D.C. Docket Nos. 08-00027-CR-3-LAC-MD, 3:08-cr-00027-LC-MD-1 UNITED STATES OF AMERICA, lllllllllllllllll Appellee, versus ROBERT L. IGNASIAK, JR., llllllllllllllllll Appellant. ________________________ Appeal from the United States District Court for the Northern District of Florida ________________________ (January 19, 2012) Before EDMONDSON, MARTIN and COX, Circuit Judges. MARTIN, Circuit Judge: The Appellant, Robert L. Ignasiak, until this case a medical doctor licensed by the State of Florida, appeals his convictions for dispensing controlled substances in violation of the Controlled Substances Act ( CSA ) and for health care fraud. Ignasiak was charged in a fifty-four count indictment with fourteen counts of health care fraud in violation of 18 U.S.C. § 1347, and forty counts of dispensing controlled substances in violation of 21 U.S.C. §§ 841(a)(1) and 841(b)(1)(C). The government s theory of prosecution for both sets of charges was substantially the same that Ignasiak had prescribed unnecessary or excessive quantities of controlled substances without a legitimate medical purpose and outside the usual course of professional practice. All fifty-four counts of the indictment related to the treatment of twenty patients. Two of the counts, twenty eight and forty eight, further charged that death resulted from the use of controlled substances prescribed by Ignasiak to two of the twenty patients, M.B. and B.E. A jury found Ignasiak guilty of forty-three of the fifty-four counts charged. He was sentenced to a total term of 292-months imprisonment, the bottom end of his advisory guideline range. Ignasiak s convictions resulted in three separate but 2 related appeals to this Court, all of which are disposed of in this opinion.1 In his merits appeal, Ignasiak claims: (1) that the evidence did not support his convictions; (2) that the District Court court abused its discretion by rejecting his proposed jury instructions on Florida law governing the use of controlled substances; (3) that the District Court abused its discretion by allowing expert witnesses to testify as to the ultimate legal issues; (4) that the District Court erred or abused its discretion by allowing the introduction of autopsy reports or handwritten medical notes without requiring testimony by their authors, in violation of the Confrontation Clause and the rules of evidence; and (5) that the District Court abused its discretion under the rules of evidence by allowing uncharged conduct to become the feature of the trial. After carefully reviewing the record and having the benefit of oral argument, we reverse Ignasiak s convictions because the admission of autopsy reports and testimony about those reports, without live in-court testimony from the medical examiners who actually performed the autopsies (and where no evidence was presented to show that the 1 Appeal No. 09-10596 is Ignasiak s merits appeal from his convictions. In Appeal No. 09-16005, Ignasiak appeals the District Court s denial of his motion to unseal the government s post-trial in camera notice to the District Court which contained impeachment evidence concerning the government s key witness, Dr. Arthur Jordan. In Appeal No. 10-11074, Ignasiak appeals the District Court s denial of his motion for new trial based upon Brady v. Maryland, 373 U.S. 83, 83 S. Ct. 1194 (1963), related to the impeachment evidence against Dr. Jordan that was the subject of the government s in camera notice. 3 coroners who performed the autopsies were unavailable and the accused had a prior opportunity to cross examine that witness), violated the Confrontation Clause under the facts of this case. Because we conclude that the fourth issue is dispositive, we decline to address the other issues raised in Ignasiak s merits appeal,2 except for the sufficiency of the evidence claim.3 While we ultimately conclude that the evidence was sufficient, the degree to which we view the government s case as less than overwhelming compels our conclusion that the Confrontation Clause violation was not harmless in this case. To give our harmful error determination sufficient context, it is necessary to describe the evidence in some detail. I. STATEMENT OF FACTS AND PROCEDURAL HISTORY During Ignasiak s nineteen day trial, the government presented forty-one witnesses, including patients, patients family members, former clinic employees, 2 See, e.g., United States v. McGough, 412 F.3d 1232, 1236 n.3 (11th Cir. 2005) (reversing convictions on one ground but declining to address other issues raised on direct appeal). 3 Even though we reverse Ignasiak s convictions on other grounds, we review his sufficiency of the evidence claim for prudential reasons. See, e.g., United States v. Bobo, 419 F.3d 1264, 1268 (11th Cir. 2005) (discussing prudential reasons for addressing sufficiency of evidence claims raised on direct appeal); United States v. Adkinson, 135 F.3d 1363, 1379 n. 48 (11th Cir.1998) ( Although not mandated by the double jeopardy clause, it is clearly the better practice for the appellate court on an initial appeal to dispose of any claim properly presented to it that the evidence at trial was legally insufficient to warrant the thus challenged conviction. ). 4 and various expert witnesses. The government also introduced exhibits including the medical charts for the twenty patients at issue in the indictment, autopsy reports, thousands of prescriptions written by Ignasiak and finally the medical charts and autopsy reports of other patients not referenced in the indictment. Until his retirement on December 15, 2005, Ignasiak operated a medical clinic in Freeport, Florida, a rural town in the Florida Panhandle where he was the only medical doctor. He had a busy medical practice and typically saw between thirty to thirty-two patients each day at fifteen minute intervals. Most patients came to renew their prescriptions, and those who worked at the clinic testified that Ignasiak always interviewed and examined his patients before they got a prescription. In addition to his regularly scheduled appointments, Ignasiak saw work-in patients, people who became sick and needed to see a doctor right away, at the rate of one or two per hour. As one former employee who worked at the clinic affirmed during the trial, if somebody was sick in Freeport and needed to see a doctor that particular day, they would come to see Dr. Ignasiak. In the spring of 2005, the federal Agency for Health Care Administration (ACHA) undertook a review of Ignasiak s files due to concern that, as a family 5 practice doctor, he was regularly billing for higher-than-normal levels of service.4 The auditor, Dr. Timothy Walker, reviewed thirty of Ignasiak s patients charts that were selected by ACHA out of more than 3,700 charts. He found that Ignasiak s charts did not justify the charges he was submitting to Medicaid because they consistently failed to document a detailed history or detailed physical exam. As Dr. Walker explained with regard to the assumption he made in reviewing the file: if it s not documented, it didn t happen. Dr. Walker was more concerned, however, about what he perceived as Ignasiak s practice of prescribing certain combinations of narcotic pain-killers in significant quantities. Of the thirty charts that he reviewed, Dr. Walker found that, aside from six children who were not receiving narcotics, most of the remaining twenty-four adults were receiving some combination of narcotic drugs. But Dr. Walker acknowledged that the charts also contained prescriptions for non-pain medication for treating illnesses such as hypertension, cholesterol, diabetes, migraines, [and] peptic ulcers. Ignasiak requested a peer meeting with Dr. Walker to review his billing. Dr. Walker agreed but informed Ignasiak that he also wanted an 4 In order to provide Medicaid services to patients, a doctor must keep written or dictated documentation of everything he does with respect to patient care, including examinations, treatment plans, patients complaints, and progress. Documentation is required because Medicaid reimburses doctors based upon the amount and complexity of the work they do. 6 explanation of Ignasiak s prescription of controlled substances. Dr. Walker did not hear back from Ignasiak By December of 2005, Ignasiak had retired and sold his Freeport medical practice to Hospital Corporations of America (HCA), which sent a replacement doctor. Dr. Maurice Marholin, a chiropractor, arrived at the clinic on January 23, 2006, following a several week period where there was no doctor present. Upon reviewing the patient files in preparation for his first visits, he was alarmed by the quantity of controlled substances being prescribed, and believed the defendant had been operating a pain management clinic rather than a family practice.5 Dr. Marholin was not equipped to run a pain management clinic and communicated his concerns to HCA, which posted a memo on the clinic s front door informing patients that narcotics would not be prescribed for two weeks. Patients who expected that Dr. Marholin would write pain medications were visibly angry when he did not.6 Dr. Marholin feared for his safety and began to wear a bullet-proof vest. 5 Dr. Marholin could not remember if he had treated any of the twenty patients at issue in the indictment. 6 Mary Bell Cosner, an HCA employee who helped with the transition, testified that when unnamed patients were unable to get their prescriptions for narcotics filled by Dr. Marholin, they became angry. 7 Dr. Marholin also noted that patients were coming from considerable distances, and he thought that [it was] unusual that people would drive past so many qualified doctors just to come to this clinic for pain management. Indeed, although the government never calculated the total number of the defendant s prescriptions, it did determine that 44,083 of them during a five-year period were for controlled substances, such as hydrocodone, alprazolam, diazepam, oxycodone, and carisoprodol. Of these 44,083 prescriptions, a relatively small number were not for Florida patients.7 According to Dr. Marholin, the medical practice rapidly declined as a result of his refusal to prescribe pain medications. In February 2006, the government seized all the patient files. Three other individuals who were involved in the transition also testified for the government. A nurse, Stephanie Hughes, who worked for Ignasiak for only six weeks, testified over objection that numerous unidentified patients came for appointments, asking for prescription refills. It was common for Ignasiak to 7 One prescription each was filled by patients who gave the filling pharmacy a zip code in Indiana, New York, and Wisconsin, three were filled by patients with a zip code in West Virginia, four were filled by patients with a zip code in Texas, six were filled by a patient with a zip code in Louisiana, seventeen were filled by patients with a zip code in Mississippi, fortyeight were filled by a patient with a zip code in Alabama, and forty-nine were filled by a patient with a zip code in Georgia. 8 prescribe Duragesic8 patches, which she testified was strange and troubled her. There was also a government expert who later testified that large doses of fentanyl are used not infrequently, for chronic, nonmalignant pain by some doctors, and that he himself ha[s] patients on fentanyl patches . . . who are clear headed and work every day. And Hughes acknowledged that no patients got prescriptions without first being examined by Ignasiak. In Hughes s experience, fentanyl patches were only written for terminally ill patients. Rebecca Clark worked briefly in Ignasiak s clinic from sometime in 2004 until he retired, and recalled the transition period. Clark testified that Ignasiak saw many patients, sometimes as many as six in an hour, and that the majority were not there because they were, in Clark s terms, sick, but rather to get monthly medication refills. A third witness, Dr. Gregory Staviski, a pain management specialist, saw approximately twenty-four of Ignasiak s patients referred to him by Dr. Marholin, and recalled that the majority wanted their medications, appeared to have little interest in being treated, and did not show up for appointments. 8 Duragesic is a transdermal patch that delivers fentanyl, a schedule II opioid pain medication, into the body slowly through the skin, where it works to relieve pain. 9 The government also offered the testimony of several patients or their relatives and various experts. Patient K.M. (counts of conviction eight, nine, twenty one, thirty eight, and forty seven) was the only patient of the twenty referenced in the indictment who testified at trial. K.M. became addicted to Lortab after another doctor prescribed it for her pain following foot surgery in 1999. Once addicted, K.M. got Lortab by buying it on the street or by telling doctors [w]hatever I needed to say to convince them to give her prescriptions. In 2001, K.M. moved to Freeport and started seeing Ignasiak. In K.M. s first visit with Ignasiak, she told him she had back problems and after a physical examination he prescribed her Lortab, Celebrex, and Skelaxin. From 2001 until May 2003, Ignasiak prescribed K.M. Lortab and various nonnarcotic medications for back and joint pain, headaches, and other non-pain related medical problems, such as congestion and allergic rhinitis. Although K.M. admitted she just wanted pain medication, she did in fact suffer from lower back pain which the pain medication relieved and allowed her to function in every day life. K.M. never told Ignasiak that she was dependent upon the pain medication and that it was only partially for pain because she was afraid to tell him the truth. According to K.M., Ignasiak examined her each time she went to see him 10 for a prescription, palpitating sore areas, having her raise and lower her arms and legs to assess the range of motion in her joints, and using a visual pain analogy scale to determine the level of her pain. K.M. stated Ignasiak was treating her as a doctor and never gave prescriptions without an exam, but there was evidence that Ignasiak refilled one to three prescriptions in between office visits over the course of four years. At one point, a pharmacist called Ignasiak to say that K.M. was attempting prematurely to fill a prescription for Lortab written by another doctor, but the defendant authorized the additional refill once it was explained that K.M. was going to the Bahamas for two weeks. K.M. testified that she did, in fact, go to the Bahamas. After this incident, Ignasiak switched K.M. s primary medication to Duragesic pain patches and required her to return the used ones to him so he could monitor how she was using the drugs. Two weeks later, K.M. returned the used patches and Ignasiak counted and inspected them, as he did every time K.M. came in for an office visit. Between 2003 2005, Ignasiak steadily increased the dosage in the patches, and continued to prescribe Lortab for breakthrough pain, and later Xanax, until his retirement. Ignasiak periodically ordered lab tests, which he reviewed with K.M. 11 and suggested she see a psychiatrist. He also cautioned K.M. only to take her medication as prescribed and not to take Xanax and Ambien together. Beverly Fein, the daughter of patient J.S. (counts of conviction thirty two and thirty seven), testified about her mother s treatments with Ignasiak from 1989 2004. J.S. did not testify. According to Fein, although J.S. had a lot of pain everywhere, Ignasiak s prescriptions to treat the pain caused her to deteriorate to the point where she stayed in her room all day watching television, frequently stumbled, slurred her words, and injured herself. The drugs made her basically non-functional. J.S. s medical records reflected that Ignasiak had scheduled repeated ultrasound tests of her gallbladder and other tests for osteoporosis, renal failure, peptic ulcer disease and hypertension, as well as referrals for her to see different specialists. In 2000, Ignasiak referred J.S. to a pain clinic but she could not afford to go. When J.S. had a double bypass surgery due to congestive heart failure, Ignasiak was the first to notice that J.S. had developed gangrene in one of her toes, and to discover that her kidneys were failing as a consequence of the surgery. Fein also testified that she told Ignasiak at one point that J.S. was addicted to her pills and had almost overdosed on several occasions. As a result, Ignasiak 12 stopped prescribing controlled substances to J.S. for awhile and stopped treating her altogether in 2004. Dana Easterly, the widower of patient B.E. (counts of conviction seven, ten, fifteen, sixteen, thirty six, and forty eight), testified that Ignasiak began treating his wife within a few months after they moved to Freeport from Louisiana in 1999. B.E. had been in a tragic car accident in 1994 in which her nine-year old daughter was killed and B.E. was ejected through the windshield head first, causing serious injuries to her face. Before seeing Ignasiak, B.E. had several reconstructive surgeries and had a history of major depression, anxiety, seizures, peptic ulcers, and fainting spells. Indeed, following the car accident, B.E. had been treated by a psychiatrist in Louisiana and had been taking Trazone, Remeron, and Valium, which made her confused and forgetful. Prior to B.E. s death on April 15, 2005, records from Sacred Heart Hospital contemporaneous with Ignasiak s treatments of B.E. showed that she was being treated for several medical conditions, many of which stemmed from the car crash, including depression and anxiety, stroke, diverticulitis, recurrent pain syndrome, chronic back pain, hepatitis, diabetes, pancreatitis, hypothyrodism, and hypertension. On her deceased daughter s birthday in 2003, her husband took her to the emergency room because she was in 13 emotional pain and could not feel the left side of her body. Hospital physicians prescribed her Xanax, Lortab, and Valium. B.E. s husband and son both testified that the prescriptions Ignasiak wrote B.E. for Lortab, Valium, Duragesic patches, and Xanax made her lethargic and nonfunctional. On the day B.E. died, B.E. seemed fine, but when her husband returned home from work she was slumped over on the sofa bed and paramedics were not able to revive her. Medical examiner Dr. Andrea Minyard conducted an autopsy three days after B.E. s death and testified that she died of multiple drug intoxication. Although Dr. Minyard could not tell what the levels of controlled substances were at the time of death from the blood tested during the autopsy, hospital records showed that the Xanax in B.E. s system was in the therapeutic range but the fentanyl was slightly higher than that. Dr. Minyard was unable to rule out the possibility that B.E. died from a stroke she suffered three weeks prior to her death. The Court admitted Dr. Minyard s autopsy report of B.E. into evidence. During Dr. Minyard s testimony, the District Court also admitted into evidence, over Ignasiak s Confrontation Clause objection,9 the autopsy reports of 9 Ignasiak interposed two objections about the autopsy reports during Dr. Minyard s testimony. Ignasiak objected at the time the autopsy reports were offered into evidence and asked to approach side bar to place his objections on the record, but was instructed to do so 14 seven other patients of Ignasiak who were not referenced in the indictment. For all seven patients, the autopsy reports concluded the cause of death was pharmaceutical drug overdoses. While Dr. Minyard had performed two of those seven autopsies, two other medical examiners, Dr. Michael Berkland and Dr. Karen L. Kelly, had performed the autopsies for the remaining five patients. during a break in the proceedings. During the break, Ignasiak objected, inter alia, because [the reports] violate the confrontation clause in the Crawford case, in that the witness is testifying from the reports; however, the individual who wrote the report is not here to testify, which inhibits effective cross-examination. Ignasiak s objections were overruled. In addition to his contemporaneous objection to the autopsy reports, Ignasiak had filed a motion in limine to exclude evidence of uncharged patients deaths, arguing that such evidence was neither relevant under Fed. R. Evid. Rule 401 nor proper under Rule 404(b). The District Court was initially inclined to grant Ignasiak s motion, but later reversed itself and granted the government s motion for reconsideration. The government s motion argued that the evidence of other deaths was relevant to all of the charges and admissible under Rule 404(b) to counter Ignasiak s defense that he was simply a small town doctor who was doing the best he could to help his patients and acted in good faith in his prescription practices. Implicit in that representation, the government argued, was Ignasiak s suggetion that [he] had no knowledge nor reason to suspect that he was improperly prescribing controlled substances or that such practices were causing death to his patients. According to the government, proof of the other deaths was relevant and important for two reasons. First, [e]vidence that during the same time periods the Defendant knew that his patients were overdosing on the medications to the point that they were dying is relevant and probative of the fact that the Defendant was not acting in good faith. Second, the government argued: a prosecution under § 841 requires proof beyond a reasonable doubt that the doctor was acting outside the bounds of professional medical practice, as his authority to prescribe controlled substances was being used not for treatment of a patient, but for the purpose of assisting another in the maintenance of a drug habit or of dispensing controlled substances for other than a legitimate medical purpose, i.e. the personal profit of the physician. Proof that the Defendant continued to prescribe massive doses of controlled substances even after he learned that his patients were dying of overdoses goes a long way to establishing that the defendant was knowingly and willfully dispensing controlled substances for other than a legitimate medical purpose, in the ordinary course of his medical practice. 15 Neither Dr. Berkland nor Dr. Kelly testified at trial. Instead, Dr. Minyard testified about the five autopsies performed by the non-testifying medical examiners, including testimony that each non-testifying medical examiner had concluded the cause of death was drug overdose for each patient, and that the manner of death was accidental for four of five patients, and suicide for the fifth patient (S.P.). In addition, Dr. Minyard testified that she agreed with the conclusions of the nontestifying medical examiners regarding cause of death for four of the five patients.10 But Dr. Minyard indicated she lacked enough information to agree or disagree with Dr. Kelly s conclusion that patient S.P. s death was a suicide. The government also presented the testimony of another medical examiner, Dr. Gary Cumberland, who had conducted an autopsy of patient M.B. (counts of conviction eighteen and twenty-eight). Dr. Cumberland testified that M.B. was a woman who looked like she had been having a downhill path [from] a medical viewpoint, [and] . . . was heading towards death. Ultimately, Dr. Cumberland concluded that M.B. died of complications from multiple drugs in her system. M.B. s toxicology report from blood drawn during the autopsy showed toxic levels of diazepam and morphine. 10 The record does not indicate Dr. Minyard s opinion on whether she agreed with the cause of death for one of the decedents. 16 Dr. Bruce Goldberger, whose laboratory completed the toxicology reports for M.B., B.E., and seven other decedents not referenced in the indictment but whose autopsy reports were admitted into evidence, testified about the toxicology as it related to each of the nine decedents. His conclusions were consistent with the cause of death listed in the autopsy reports. In addition, the government offered expert testimony concerning the medical propriety of Ignasiak s treatments. Pharmacologist Paul Doering reviewed the files of twenty of Ignasiak s patients, with a focus on the prescriptions issued to those patients. Dr. Doering testified that certain combinations of drugs Valium, Duragesic, and Soma may have a potentially dangerous synergistic effect when used together, that the defendant prescribed such combinations to a number of patients, and that this caused Dr. Doering serious[] concern. Although Dr. Doering was not a physician and therefore would not have diagnosed patients, his review of the twenty patient files corresponding to the indictment caused him concern because the defendant had a pattern of prescribing certain drug combinations that were unhealthy and moreover because it didn t look like a whole lot of good was going to come to those combinations. While criticizing Ignasiak s long term use of opiods to treat pain, Dr. Doering also acknowledged that such use had been approved by the 17 Federal Drug Administration and that it was difficult to determine when patients were faking their pain. Dr. Arthur Jordan, a South Carolina medical doctor who specializes in pain management, was the centerpiece of the government s presentation regarding prescribing practices. He testified for almost three days concerning his opinion on the generally accepted treatment philosophy in the field of pain management; the dangers and risks of using narcotics individually or in combination with other drugs; the critical importance of documenting a patient s file regarding complaints, care, and treatment; the warning signs of prescription drug abuse; and the responsibility of medical doctors to ensure controlled substances are properly used and prescribed. But Dr. Jordan was not familiar with Florida law and practices concerning pain management and the practice of medicine. Dr. Jordan reviewed the medical and pharmacy records of the twenty patients named in the indictment. Dr. Jordan offered the opinion that the documentation in Ignasiak s charts did not support the quantities and substances prescribed. For each of the patients referenced in the indictment, Dr. Jordan affirmed or stated, over defense counsel s objection, that Ignasiak had not prescribed the controlled substances for a legitimate medical purpose or in the usual course of medical practice. Dr. Jordan also testified that M.B. and B.E. 18 might not have died if they had not been prescribed controlled substances by Ignasiak. Then on cross-examination, Dr. Jordan acknowledged that, due to the serious medical conditions reflected in the records, a number of Ignasiak s treatment practices were reasonable, legitimate, appropriate, the conduct of a doctor and/or in the usual practice of medicine. At least for some of Ignasiak s patients, Dr. Jordan acknowledged that it may have been reasonable to treat them with controlled substances, including pain medication, on some occasions. Also, in at least some instances, the records reviewed by Dr. Jordan also reflected that other health care providers had prescribed the same, or similar, controlled substances to the same patients. Furthermore, while Dr. Jordan generally faulted Ignasiak for inadequately documenting his patients charts, he recognized that some of the charts showed Ignasiak had counseled his patients on health issues, ordered diagnostic testing, and referred patients to various specialists for further treatment, including pain specialists. The government also introduced the medical records of eight patients not mentioned in the indictment. These records contained handwritten notations about reports mainly from unidentified or anonymous people claiming that the patients were selling or otherwise abusing their medications. Further, the government 19 introduced sixty-one Drug Utilization Review letters that Ignasiak received from the State of Florida between 1996 and 2006.11 Ignasiak responded to only one of these letters. Although Ignasiak did not testify, the defense did call its own expert witnesses to dispute the government s allegations. The defense called pharmacology expert Leonard Rappa, who testified that the combinations of drugs prescribed by Ignasiak were not excessive, either individually or in combination, and were appropriate for the conditions documented in the medical records from a pharmaceutical standpoint. He also testified that Ignasiak s prescriptions never exceeded the diagnosis-specific dosages and quantities listed in the Physician s Desk Reference ( PDR ), a compilation of all medications, monographs, and FDA approval limitations. The government did not dispute this last point. Dr. David Fowler, the Chief Medical Examiner for the State of Maryland, reviewed the autopsy reports that were admitted during the trial. He criticized the delays in the autopsies and testified that toxicology reports can be misleading due to the redistribution of drugs in the body after death. Based on a review of the 11 Medicaid has a provider review system in which a reviewing pharmacist or physician may, upon identifying a concern such as a an overutilization of a product or a duplication in therapy, request that the provider be sent a Drug Utilization Review (DUR) letter advising the provider of the concern. 20 autopsy reports, Dr. Fowler opined that the deaths were either natural, the result of suicides, or uncertain. At the close of the evidence, the District Court denied Ignasiak s motion for judgment of acquittal. The jury returned a verdict of guilty as to forty three of the counts and a verdict of not guilty as to eleven counts pertaining to five patients. Ignasiak timely appealed. II. STANDARDS OF REVIEW We review the sufficiency of the evidence de novo, viewing the evidence and all reasonable inferences and credibility choices in favor of the government and the jury s verdict. United States v. Friske, 640 F.3d 1288, 1290 91 (11th Cir. 2011). A conviction must be affirmed unless there is no reasonable construction of the evidence from which the jury could have found the defendant guilty beyond a reasonable doubt. United States v. Garcia, 405 F.3d 1260, 1269 (11th Cir. 2005). A defendant s claim that his Sixth Amendment rights were violated is reviewed de novo. United States v. Gari, 572 F.3d 1352, 1361 62 (11th Cir. 2009). III. DISCUSSION A. Sufficiency of the Evidence 21 As noted above, Ignasiak was charged in fourteen counts of health care fraud and forty counts of dispensing controlled substances. The government s theory of prosecution for both sets of charges was substantially the same that Ignasiak had prescribed unnecessary or excessive quantities of controlled substances without a legitimate medical purpose and outside the usual course of professional practice. Thus, although Ignasiak was convicted of both substantive fraud counts and dispensing controlled substances counts, the convictions are inextricably intertwined. The indictment compels this conclusion because the manner and means in which the fraudulent scheme was carried out directly related to whether Ignasiak s prescriptions for controlled substances were legitimate and within the usual course of professional practice. Indeed, the government characterized the fourteen fraud counts as a scheme to defraud health care providers . . . by billing for office visits when they were, in reality not more than drug distribution. Thus, we evaluate the sufficiency of the evidence for the fraud and dispensing counts together in this case, focusing upon the requirements under the CSA. The CSA comprehensively regulates the flow of certain classes of drugs in the United States, ranging from common sleep aids to crack cocaine. The statute, which is administered by the Drug Enforcement Agency ( DEA ), provides for the 22 scheduling, rescheduling, and descheduling of controlled substances into one of five categories based on various factors, such as their addictive properties and risk of side effects. 21 U.S.C. §§ 811, 812. It criminalizes the manufacture, distribution, dispensation, or possession of these substances, except in a manner authorized by the CSA. 21 U.S.C. §§ 841(a)(1), 844(a). One use authorized under the statute is for medical treatment by a physician registered with the DEA pursuant to his state medical license. See id. § 823(f). A DEA-registered physician may dispense controlled substances so long as he does so for a legitimate medical purpose and while acting in the usual course of his professional practice. 21 U.S.C. §§ 830(b)(3)(A)(ii); accord id. at 802(21); 21 CFR § 1306.04(a) (2005). Conversely, [t]o convict [a DEA-licensed physician under § 841], it [is] incumbent upon the government to prove that he dispensed controlled substances for other than legitimate medical purposes in the usual course of professional practice, and that he did so knowingly and intentionally. United States v. Guerrero, 650 F.2d 728, 730 (5th Cir. July 16, 1981) (citation omitted).12 12 In Bonner v. City of Prichard, 661 F.2d 1206, 1209 (11th Cir. 1981) (en banc), this Court adopted as binding precedent all decisions of the former Fifth Circuit handed down prior to October 1, 1981. 23 About the sufficiency of the evidence, there was inculpating evidence of a circumstantial nature that Ignasiak was prescribing controlled substances outside the usual course of professional practice. Indeed, there is evidence in the record showing that Ignasiak sometimes failed to ask for explanations of the erratic behavior of certain patients, and did not always exercise proper precautions when informed by patients that they feared addiction to the narcotics they were prescribed. Dr. Marholin testified that Ignasiak s patients expressed anger when told they could no longer receive narcotics as a matter of course. Dr. Staviski testified that at least some of Ignasiak s patients lacked any apparent desire to recover from their purported illnesses. But ample record evidence also suggests that Ignasiak s practice exposed him to sick patients afflicted by legitimately painful conditions. Indeed, the medical records of all twenty patients referenced in the indictment document illnesses or conditions that caused them pain, anxiety and/or depression, ailments that could well have justified the use of controlled substances within the range of discretion accorded physicians. It was also not disputed at trial that all of Ignasiak s prescriptions were written within the diagnosis-specific guidelines contained in the Physician s Desk Reference. 24 The government relied largely upon medical records at trial. Specifically, of the twenty patients charged in the indictment, the government conducted independent examinations of only two the autopsies of B.E. and M.B. Further, the government called only one patient referenced in the indictment, K.M., to testify. Rather than calling or examining the remaining seventeen patients named in the indictment, the government relied almost exclusively on the testimony of one doctor Dr. Jordan who testified extensively about his review of Ignasiak s charts concerning those charged patients. Dr. Jordan did not interview or examine a single patient to determine whether their treatments were warranted. Based upon this review, we conclude that, when viewed in the light most favorable to the government, the evidence was sufficient to support Ignasiak s conviction. In this case, like in Merrill, the government introduced evidence that Ignasiak had written more than 43,000 prescriptions for controlled substances over a five year period, which it then alleged was conduct outside the usual course of professional practice. 513 F.3d at 1299 1301. Based upon this evidence, in combination with the experts testimony, we cannot say the jury could not have found the defendant guilty under any reasonable construction of the evidence. Id. at 1299. Further, the fact the jury acquitted Ignasiak of several counts involving patients who did not testify while convicting him on others, 25 sufficiently minimizes the risk identified in Cuong that Ignasiak s jury merely convicted him based only upon guilt by association. Cuong, 18 F.3d at 1142. B. Autopsy Reports Ignasiak argues that the District Court violated the Confrontation Clause by allowing the government to introduce, and Dr. Minyard to testify about, five autopsy reports which neither she nor any other witness had authored.13 During Ignasiak s trial, the government introduced the autopsy reports of five of Ignasiak s former patients in which the cause of death was determined to be, at least in part, intoxication from controlled substances. In each instance, the government offered these autopsy reports through Dr. Minyard in her capacity as chief medical examiner and records custodian for the Office of the Medical Examiner, District I. Because Dr. Minyard had not performed the autopsies, defense counsel objected based upon the Confrontation Clause. The District Court overruled Ignasiak s objection. As a result, the five autopsy reports were admitted into evidence and Dr. Minyard was allowed to testify as to their contents with respect to each patient s death having been caused by controlled substances.14 13 Ignasiak also argues that his Confrontation rights were violated because anonymous notations on his patients charts were admitted into evidence over his objection. Because we find the autopsy issue dispositive, we need not address this aspect of his claim. 14 The autopsy reports appear to have been admitted into evidence under the hearsay exception for documents kept in the regular course of business under Fed. R. Evid. 803(6). 26 The government argues that the autopsy reports admitted through the custodian of records as business records are not subject to the Confrontation Clause s rigor. The government argues alternatively that, even assuming error occurred, it was harmless because the reports did not concern patients referenced in the indictment. To resolve this dispute, we must first determine whether the autopsy reports at issue are testimonial evidence subject to the Confrontation Clause. For the following reasons, we conclude that the answer to this question is yes. In Crawford v. Washington, 541 U.S. 36, 68 69, 124 S. Ct. 1354, 1374 (2004), the Supreme Court wrote, [w]here testimonial statements are at issue, the only indicium of reliability sufficient to satisfy constitutional demands is the one the Constitution actually prescribes: confrontation. This means that the prosecution may not introduce testimonial hearsay against a criminal defendant, regardless of whether such statements are deemed reliable, unless the defendant Under that rule, the availability of the declarant is immaterial. Of course, admissibility under the rules of evidence does not vitiate Confrontation Clause requirements. See Crawford v. Washington, 541 U.S. 36, 50 51, 124 S. Ct. 1354, 1364 (2004) ( [W]e . . . reject the view that the Confrontation Clause applies of its own force only to in-court testimony, and that its application to out-of-court statements introduced at trial depends upon the law of Evidence for the time being. Leaving the regulation of out-of-court statements to the law of evidence would render the Confrontation Clause powerless to prevent even the most flagrant inquisitorial practices. (citations omitted)). The record is unclear as to whether Dr. Kelly or Dr. Berkland were unavailable at the time of trial. Nor is there any indication that Ignasiak had a prior opportunity to cross-examine them. 27 has an opportunity to cross-examine the declarant, or unless the declarant is unavailable and the defendant had prior opportunity for cross-examination. Id. at 53 54, 68, 124 S. Ct. at 1365 66, 1374. Crawford described the class of testimonial statements covered by the Confrontation Clause as follows: Various formulations of this core class of testimonial statements exist: ex parte in-court testimony or its functional equivalent-that is, material such as affidavits, custodial examinations, prior testimony that the defendant was unable to cross-examine, or similar pretrial statements that declarants would reasonably expect to be used prosecutorially; extrajudicial statements ... contained in formalized testimonial materials, such as affidavits, depositions, prior testimony, or confessions; statements that were made under circumstances which would lead an objective witness reasonably to believe that the statement would be available for use at a later trial. Crawford, at 51-52, 124 S. Ct. 1364 (quotation marks and citations omitted). Forensic reports constitute testimonial evidence. To wit, in Melendez-Diaz v. Massachusetts, ---U.S.---, 129 S. Ct. 2527 (2009), the Supreme Court held that a forensic laboratory report stating that an unknown substance was cocaine qualifies as testimonial evidence to which the Confrontation Clause applies. Id. at 2532. The Court reasoned that because the certificates are incontrovertibly a solemn declaration or affirmation made for the purpose of establishing or proving some fact, they are functionally identical to live, in-court testimony, doing precisely what a witness does on direct examination i.e. offering proof that the substance 28 was cocaine. Id. (quotation marks and citations omitted). As a result, and because scientific evidence is no more neutral or reliable than other testimonial evidence, confrontation serves to ensure its accuracy by weed[ing] out not only the fraudulent analyst, but the incompetent one as well. Id. at 2537. Moreover, the scientific nature of forensic reports does not justify subjecting them to lesser scrutiny than other testimonial evidence. To the contrary, in Bullcoming v. New Mexico, ---U.S.---, 131 S. Ct. 2705 (June 23, 2011), the Court made clear that the Sixth Amendment requires that, when introducing testimonial forensic evidence, the prosecution must present testimony by a scientist who was actually involved in preparing that forensic evidence. Id. at 2710, 2713. In so doing, the Court specifically rejected the use of so-called surrogate testimony, which in Bullcoming was that of a colleague from the same lab that prepared the disputed forensic report, but who had not specifically worked on the reports in question. Id. at 2710, 2712 13. Even though the colleague was able to testify as to the efficacy and reliability of the laboratory equipment, and also whether normal protocol was followed, the Court explained that the comparative reliability of an analyst s testimonial report drawn from machineproduced data does not overcome the Sixth Amendment bar . . . [because] the obvious reliability of a testimonial statement does not dispense with the 29 Confrontation Clause. Id. at 2714. Instead, only testimony by the actual scientist who prepared the forensic report could provide insight into the particular test and testing process . . . employed, and also expose any lapses or lies on the certifying analyst s part. Id. Applying the reasoning of Crawford, Melendez-Diaz, and Bullcoming, we conclude that the five autopsy reports admitted into evidence in conjunction with Dr. Minyard s testimony, where she did not personally observe or participate in those autopsies (and where no evidence was presented to show that the coroners who performed the autopsies were unavailable and the accused had a prior opportunity to cross examine them), violated the Confrontation Clause.15 For 15 Even before Bullcoming many states which had considered the issue concluded that autopsy reports are testimonial and subject to the Confrontation Clause. See Smith v. Alabama, 898 So. 2d 907, 917 (Ala. Crim. App. 2004) (holding that admission of autopsy evidence and reports, without testimony of medical examiner who performed autopsy, violated defendant s Sixth Amendment right to confrontation but was harmless error); Commonwealth v. Avila, 912 N.E. 2d 1014, 1029 (Mass. 2009) (holding that the autopsy report was testimonial hearsay); Cuesta-Rodriguez v. State, 241 P.3d 214, 228 (Okla. Crim. App. 2010) (holding that admission of autopsy report was testimonial); State v. Locklear, 681 S.E. 2d 293, 305 (N.C. 2009) (holding that the trial court erred under Crawford by admitting forensic analyses performed by a forensic pathologist and a forensic dentist who did not testify ); Wood v. Texas, 299 S.W. 3d 200, 209 10 (Tex. Crim. App. 2009) (holding the autopsy report was testimonial where police suspected the death was a homicide); but see People v. Cortez, 931 N.E. 2d 751, 756 (Ill. App. Ct. 2010) (holding that autopsy reports are non-testimonial business records). Furthermore, although the First Circuit held United States v. Feliz, 467 F.3d 227 (1st Cir. 2006) that an autopsy report is admissible as a business record, id. at 236 37, the Feliz case came before Melendez-Diaz, which as discussed below rejected that same business record argument as applied to the forensic evidence at issue in that case. See 129 S. Ct. at 2538. As such, we conclude that Feliz has little persuasive value on this issue. 30 background, see Melendez-Diaz, 129 S. Ct. at 2538 (business record exception does not encompass documents generated by an entity that regularly produc[es] . . . evidence for use at trial ). We think the autopsy records presented in this case were prepared for use at trial. Under Florida law, the Medical Examiners Commission was created and exists within the Department of Law Enforcement. Fla. Stat. § 406.02. Further, the Medical Examiners Commission itself must include one member who is a state attorney, one member who is a public defender, one member who is sheriff, and one member who is the attorney general or his designee, in addition to five other non-criminal justice members. Id. The medical examiner for each district shall determine the cause of death in a variety of circumstances16 and shall, for that purpose, make or have performed such examinations, investigations, and autopsies as he or she shall deem necessary or as shall be requested by the state attorney. 16 Those circumstances include when any person dies in the state of Florida: (1) Of criminal violence; (2) By accident; (3) By suicide; (4) Suddenly, when in apparent good health; (5) Unattended by a practicing physician or other recognized practitioner; (6) In any prison or penal institution; (7) In police custody; (8) In any suspicious or unusual circumstance; (9) By criminal abortion; (10) By poison; (11) By disease constituting a threat to public health; (12) By disease, injury, or toxic agent resulting from employment; [or] (b) When a dead body is brought into the state without proper medical certification; [or] (c) When a body is to be cremated, dissected, or buried at sea. Fla. Stat. § 406.11)(1)(a) (c). 31 Fla. Stat. § 406.11(1). Further, any person who becomes aware of a person dying under circumstances described in section § 406.11 has a duty to report the death to the medical examiner. Id. at § 406.12. Failure to do so is a first degree misdemeanor. Id. Upon receipt of such notification . . . the district medical examiner . . . shall examine or otherwise take charge of the dead body and shall notify the appropriate law enforcement agency. Fla. Stat. § 406.13. Then, after the cause of death is determined, the medical examiner is required to report or make available to the state attorney, in writing, her . . . determination as to the cause of death. Id. The medical examiner may retain [a]ny evidence or specimen coming into the possession of said medical examiner in connection with any investigation or autopsy, or deliver it to law enforcement. Id. Likewise, law enforcement has a duty to make [a]ny evidence material to the . . . cause of death in the possession of law enforcement available to the medical examiner. Fla. Stat. at 406.14.17 In light of this statutory framework, and the testimony of Dr. Minyard, the autopsy reports in this case were testimonial: made under circumstances which would lead an objective witness reasonably to believe that the statement would be 17 That medical examiners worked closely with law enforcement in connection with this case is further illustrated by the fact that Dr. Minyard herself testified that she relied upon information collected by deputies on the scene in her investigation of B.E. 32 available for use at a later trial. 18 United States v. Baker, 432 F.3d 1189, 1203 (11th Cir. 2005) (quotation marks omitted). As such, even though not all Florida autopsy reports will be used in criminal trials, the reports in this case are testimonial and subject to the Confrontation Clause. See United States v. Caraballo, 595 F.3d 1214, 1228 29 (11th Cir. 2010) (finding admission without presentation of original witness of government form that had non-testimonial primary purpose did not violate Confrontation Clause). We add that there are additional reasons to view these autopsy reports as testimonial and therefore subject to the requirements of the Confrontation Clause. Medical examiners are not mere scriveners reporting machine generated raw-data. See Bullcoming,131 S. Ct. at 2714 (rejecting argument that laboratory testing analyst is mere scrivener simply transcribing machine generated results and therefore not the true accuser for Confrontation Clause purposes (citations omitted)). It is clear from our review of Dr. Minyard s testimony concerning the autopsy she performed on B.E., as well as our review of the autopsy reports at issue, that the observational data and conclusions contained in the autopsy reports are the product of the skill, methodology, and judgment of the highly trained 18 In this case, Dr. Minyard s testimony affirms that an autopsy, as performed by a medical examiner, is a forensic test. 33 examiners who actually performed the autopsy.19 The autopsy report for T.S. illustrates the point. T.S. s autopsy was performed by associate medical examiner Dr. Karen Kelly, a medical doctor. We must accept Dr. Kelly s autopsy report as accurate, since she did not testify and was not cross-examined. With that in mind, we observe that Dr. Kelly followed a comprehensive autopsy protocol, which included a detailed external and internal examination of the body. Her autopsy report states her observations and impressions of T.S. s respiratory, cardiovascular, gastrointestinal, genitourinary, nervous, musculoskeletal systems, as well microscopic descriptions of T.S. s liver, coronary arteries, ventricle, lung, kidney, and thyroid. And most relevant to Ignasiak s case, Dr. Kelly s autopsy report states that T.S. s cause of death was combined toxic effects of ethanol, alprazolam, hydrocodone and carisprodol, and that the manner of death was by accident. In short, there is little, if any, raw data or conclusions reflected in T.S. s autopsy report aside from the results of toxicology testing that is not dependent 19 Dr. Minyard s testimony illustrates that discretion, judgment, and direct observation are critical to the medical examiners investigation of the cause of death. For example, in deciding whether to preserve tissue samples of the brain for later microscopic inspection for evidence of stroke, Dr. Minyard testified that she relies upon her judgment and direct observations during the autopsy. Rather than relying upon diagnostic tools such as CT scan or an MRI, Dr. Minyard stated: I m going to use my eyes, and I m going to use my hands to figure out what caused the death. 34 upon the skill, methodology and judgment exercised by the actual medical examiner who performed the autopsy.20 As T.S. s autopsy report illustrates, the reports in this case are replete with the extensive presence and intervention of human hands and exercise of judgment that presents a risk of error that might be explored on cross-examination. Melendez-Diaz, 129 S. Ct. at 2537. In this way, these autopsy reports are like many other types of forensic evidence used in criminal prosecutions. They may be invalid or unreliable because of the examiner s errors, omissions, mistakes, or bias. To paraphrase the Supreme Court s reasoning in Melendez-Diaz, there is little reason to believe that confrontation will be useless in testing [medical examiners s] honesty, proficiency, and methodology the features that are commonly the focus in the crossexamination of experts. Id. at 2538. Our conclusion that the autopsy reports in this case are testimonial compels the rejection of Dr. Minyard s live in-court testimony as a constitutionally 20 We express no opinion about whether the toxicology results themselves present Confrontation Clause problems because that issue was not raised by Ignasiak and, in any event, Dr. Goldberger actually testified at trial as to the results of his laboratory s toxicology testing. However, that the toxicology results were subject to cross-examination does not vitiate the confrontation problem created by the admission of the autopsy reports. As Dr. Goldberger explained during his testimony, toxicology results alone are not definitive of the cause of death: You wouldn t want to use the report alone to certify the cause and manner of death. It s required, but the medical examiner has to take under consideration the remainder of the findings, obviously, including from the autopsy and the investigators involved. The report alone is not sufficient. 35 adequate surrogate for the actual medical examiner who performed the autopsy. Although Dr. Minyard was qualified as an expert, there is no evidence that she observed the autopsies in question. Therefore, Dr. Minyard is in precisely the same position as the surrogate whose testimony was rejected in Bullcoming. Admission of the autopsy reports on solely this testimony absent evidence that the actual medical examiners who performed the autopsy were unavailable and the accused had a prior opportunity to cross examine them thus violated Ignasiak s Sixth Amendment rights under the Confrontation Clause.21 Further, in addition to the fact that the autopsy reports were admitted into evidence, Ignasiak suffered prejudice because of the significant role that the nontestifying experts played in conducting the autopsies and generating the data contained in the autopsy reports upon which Dr. Minyard based her opinion. That 21 We are aware that the Supreme Court recently granted certiorari in People v. Williams, 939 N.E. 2d 268 (Ill. 2010), cert. granted, Williams v. Illinois, ---U.S.---, 131 S. Ct. 3090 (June 28, 2011) (No. 10-8505), to decide the following question presented: [w]hether a state rule of evidence allowing an expert witness to testify about the results of DNA testing performed by non-testifying analysts, where the defendant has no opportunity to confront the actual analysts, violates the Confrontation Clause. See http://www.supremecourt.gov/Search.aspx?FileName=/ docketfiles/10-8505.htm. But Ignasiak s case is materially different than Williams. Williams involves admission of independent expert witness opinion about DNA testing performed by a non-testifying witness, but neither the non-testifying analysts report was admitted into evidence nor did the testifying expert read the contents of that report into evidence. See People v. Williams, 939 N.E. 2d 268, 345 (Ill. 2010). Here, in contrast, the autopsy reports themselves were admitted into evidence and Dr. Minyard essentially read the contents of these reports into evidence. Moreover, the manner in which Dr. Minyard testified about the medical examiners s conclusions leaves us unable to view her brief expression of agreement with the non-testifying medical examiners conclusions as truly independent expert opinion. 36 Dr. Minyard may have briefly expressed her own independent agreement with the non-testifying medical examiner s conclusions regarding cause of death only compounded the Confrontation Clause error that occurred. Because Dr. Minyard had neither performed nor been present during the autopsies in question, she was not in a position to testify on cross-examination as to the facts surrounding how the autopsies were actually conducted or whether any errors, omissions, or mistakes were made.22 See Bullcoming, 131 S. Ct. at 2715 ( But surrogate testimony of the kind [the testifying witness] was equipped to give could not convey what [the testing analyst] knew or observed about the events his certification concerned, i.e., the particular test and testing process he employed. Nor could such surrogate testimony expose any lapses or lies on the certifying analyst s part. (footnotes omitted)). Yet, Dr. Minyard gave her professional imprimatur to this effect. That Dr. Minyard exacerbated the Confrontation Clause violation is also evident from the limits of her testimony. For example, without having done or observed the autopsy, Dr. Minyard could not testify from direct knowledge about the condition of a particular patient s heart, lungs or brain and, as a result, whether 22 Indeed, this point is demonstrated by Dr. Minyard s testimony when she affirmed that she did not know if Dr. Berkland had made a mistake in reporting his observations about A.B. s lividity in his autopsy report. 37 that patient may have actually died from a heart attack, stroke, or some cause other than drug overdose. To answer that question on cross-examination, Dr. Minyard would have to refer to Dr. Kelly s and Dr. Berkland s autopsy reports. But as already noted the ultimate conclusions and supporting findings reflected in the autopsy reports are the product of the examiner s skill and judgment, not an infallible machine that requires no human intervention. Again, as Melendez-Diaz instructs, because human judgment and skill were involved, we cannot assume the non-testifying medical examiner s findings were reliable even if the examiners themselves possessed the scientific acumen of Mme. Curie and the veracity of Mother Teresa. Melendez-Diaz, 129 S. Ct. at 2537 n.6. Dr. Minyard s ostensibly independent opinion, therefore, cannot truly be regarded as independent in a way that is meaningful under Melendez-Diaz and Bullcoming, which emphasize the absolute right to confront the analyst, in this case the medical examiner, who actually performed the test. Because we determine that Ignasiak s Sixth Amendment rights were violated, we must consider whether the error was harmless. United States v. Gari, 572 F.3d 1352, 1362 (11th Cir. 2009). The test for determining whether a constitutional error is harmless is whether it appears beyond a reasonable doubt that the error complained of did not contribute to the verdict obtained. Id. 38 (quotation marks omitted). Although the reports referred to patients not referenced in the indictment, we cannot ignore the powerful impact this collateral evidence must have had on Ignasiak s jury. Here, the jury heard live in-court testimony from two different medical examiners, including Dr. Minyard, that two patients referenced in the indictment M.B. and B.E. died as a result of intoxication from controlled substances given to them by Ignasiak. But the jury was also permitted to consider testimonial evidence from two non-testifying medical examiners, Dr. Berkland and Dr. Kelly, by the admission of their autopsy reports, that five other patients of Dr. Ignasiak died from prescription drug overdoses. Based upon our review of the entire record in this case, we cannot say beyond a reasonable doubt that admission of the five autopsy reports did not contribute to all of the verdicts obtained. To be sure, each of the counts charged in the indictment represented individual events of either fraud or dispensing controlled substances. But all of the fraud counts shared a common denominator with the controlled substances counts: the government s overarching theory of prosecution that Ignasiak had prescribed unnecessary or excessive quantities of controlled substances without a legitimate medical purpose and outside the usual 39 course of professional practice. 23 In this way, both the substantive fraud and dispensing controlled substances convictions were inextricably intertwined and directly related to Dr. Ignasiak s good faith defense. It bears repeating that, over Ignasiak s objections, the government introduced evidence of other patients deaths by successfully arguing that this evidence was relevant to all of the charges and admissible under Rule 404(b) to counter Ignasiak s defense that he was simply a small town doctor who was doing the best he could to help his patients and acted in good faith in his prescription practices. Implicit in that representation, the government argued, was Ignasiak s suggetion that [he] had no knowledge nor reason to suspect that 23 During closing argument, the government reviewed the fraud counts of the indictment with the jury and then expressly tied them to the legitimacy of Ignasiak s prescription practices for controlled substances as follows: More specifically, as to Counts 1 through 14, the scheme to defraud. A scheme to defraud basically states, and the indictment states, that he engaged in a scheme to defraud a health care benefit program and to obtain, by materially false and fraudulent pretenses, representations and promises, health care benefit program money -- in other words, the Medicaid information that you heard during the trial, the Blue Cross information that you heard during the trial, the other insurance money that you heard during the trial, is all money that he gained because of this scheme to defraud. And that s fancy legal terms, basically, for that this doctor did what he did for something other than legitimate purpose, and to defraud the health care benefit programs out of money, out of the money for the prescriptions that were not legitimately prescribed and out of the office visits that were not medically necessary office visits. 40 he was improperly prescribing controlled substances or that such practices were causing death to his patients. According to the government, proof of the other deaths was relevant and important for two reasons: (1) [e]vidence that during the same time periods [Ignasiak] knew that his patients were overdosing on the medications to the point that they were dying is relevant and probative of the fact that [Ignasiak] was not acting in good faith ; (2) [p]roof that [Ignasiak] continued to prescribe massive doses of controlled substances even after he learned that his patients were dying of overdoses goes a long way to establishing that the defendant was knowingly and willfully dispensing controlled substances for other than a legitimate medical purpose, in the ordinary course of his medical practice. 24 The government returned to this theme during its closing argument, 24 That the uncharged patients deaths were considered material to the jury s deliberations is also demonstrated by the government s mid-trial request to excuse one of the seated jurors in light of that juror s attitude toward suicide. Initially, we would note that the issue of suicide came up, apparently, in response to testimony about whether one of the uncharged patient s death was caused by suicide and was therefore not accidental, e.g., Dr. Berkland s autopsy report for B.D. characterized her death as a suicide, but Dr. Minyard refused on crossexamination to agree or disagree with this conclusion. Shortly after the suicide issue came up again during the testimony of Dr. Goldberger, the District Court announced during a break that it had learned through a court security officer that one of the jurors had a concern about continuing to serve on the jury. While questioned individually by the Court in the presence of the parties, the juror disclosed that she was bothered to hear about suicides or attempts to commit suicide because her son had attempted suicide on a number of occasions. The juror further explained that when an individual commits suicide it is that person s choice and that she could not blame another for his or her suicide. Thereafter, the government requested that the juror be excused based on her inability to fairly and impartially view the evidence with regard to the potential suicides, although the government acknowledged [t]here is not a suicide involved in the actual charges. 41 arguing that the deaths of uncharged patients put [Ignasiak] on notice that his prescribing habits could be killing people, and that at the very least should have served as notice to Dr. Ignasiak that perhaps there was something wrong with the way that he was prescribing [controlled substances]. In light of the government s stated purposes for admitting the uncharged patients deaths at trial, we cannot conclude beyond a reasonable doubt that the evidence did not achieve its intended purpose and contribute to the verdicts obtained. In this case, the government s evidence was not overwhelming. While the government s evidence was legally sufficient to support the conviction on appeal, a reasonable jury could have acquitted based on all of the evidence. Our conclusion that the evidence supports the convictions in this case follows from the requirement that we review the evidence in the light most favorable to the prevailing party. See Merrill, 513 F.3d at 1299. Having arrived at this conclusion In response to the District Court s question, [w]hat does that legally have to do with your case? , the government explained, inter alia: Well, . . . if the defense is going to argue that these [uncharged patient deaths], instead of having accidently overdosed on drugs, actually committed suicide, then this juror has indicated that this is an emotional response that she is incapable of putting any kind of responsibility on anybody but the patient, which is contrary to the evidence, and shows an inability to be able to fairly and impartially determine whether or not these individuals actually did commit suicide. Ignasiak opposed discharging the juror, arguing, inter alia, that all this deals with the 404(b) evidence, which is only going to be used by the jurors to determine lack of mistake or knowledge or intent. The juror was ultimately removed from further service from the jury. 42 regarding the sufficiency of the evidence by this route, however, we must be less tolerant of the idea that errors committed during the trial of this case are acceptable because they are harmless. See Strickland v. Washington, 466 U.S. 668, 696, 104 S. Ct. 2052, 2069 (1984) ( a verdict or conclusion only weakly supported by the record is more likely to have been affected by errors than one with overwhelming record support ). We cannot conclude that the government, as beneficiary of the constitutional error, proved beyond a reasonable doubt that admitting evidence of five more patients who allegedly died as a result of Ignasiak s conduct did not contribute to the verdict obtained. Chapman v. California, 386 U.S. 18, 24, 87 S. Ct. 824, 828 (1967). IV. NEW TRIAL AND RECORD SEALING APPEALS In light of our ruling, Ignasiak s separate appeal in case number 10-11074 from the denial of his motion for new trial is moot. This is not the case, however, for Ignasiak s appeal in case number 09-16005. In that separate proceeding, Ignasiak appeals the District Court s denial of his motion to unseal the government s post-trial in camera notice to the District Court, which contained impeachment information about the government s key witness, Dr. Arthur Jordan. In light of our serious concerns about the government s reasons for keeping that notice sealed, we reverse the District Court s denial of Ignasiak s motion. 43 To briefly provide background, several months after Ignasiak s conviction, the government filed a pleading under seal in the District Court entitled Government s In Camera Notice to the Court (the Notice ), along with an affidavit from one of the trial prosecutors, Assistant United States Attorney ( AUSA ) Benjamin Beard. The Notice revealed for the first time that Dr. Jordan engaged in criminal conduct beginning at an unspecified time up to and continuing until 2006. Specifically, Dr. Jordan had, on nine separate occasions, used a counterfeit badge and his United States Marshal credentials to pose as an on-duty U.S. Marshal in order to carry firearms on commercial airplanes while on personal travel. On the ninth flight, a Transportation and Security Administration ( TSA ) agent discovered Dr. Jordan s ploy, and seized the weapons, counterfeit badge, and Marshal Service credentials. The South Dakota U.S. Attorney s Office opened an investigation of Dr. Jordan. Although Dr. Jordan had engaged in similar criminal conduct at least eight times before, thereby committing multiple violations of 18 U.S.C. §§ 912 and 1001 and 49 U.S.C. § 46505, the South Dakota U.S. Attorney allowed Dr. Jordan to enter into a pre-trial diversion agreement in which Dr. Jordan paid $2,000 and agreed not to carry any concealed weapons except while on official business. 44 The Notice also set out the government s position that the failure to disclose the information did not violate Ignasiak s constitutional rights under Brady v. Maryland, 373 U.S. 83, 83 S. Ct. 1194 (1963) and Giglio v. United States, 405 U.S. 150, 92 S. Ct. 763 (1972). Specifically, the Notice averred that Ignasiak could not make a successful Brady claim because the prosecutor did not personally know about Dr. Jordan s conduct, or the plea agreement, during Ignasiak s prosecution. Shortly after the government filed the Notice, Ignasiak filed a motion to unseal it together with the affidavit. He argued that allowing the government to unilaterally file its notice under seal violates his Fifth Amendment, common law, and First Amendment rights of access to court proceedings and documents. The government filed a response urging against disclosure. In it, the government acknowledged that ordinarily there is a presumption of openness in legal proceedings, but asserted that in this case the District Court should keep the Notice sealed, so as to protect Dr. Jordan s privacy and protect him from any potential retaliation. These concerns, the government submitted, warranted sealing the information about Dr. Jordan s previous conduct. Indeed, [o]n balance, if the undersigned erred in filing under seal, no prejudice has been caused anyone, including the witness; whereas, if it had been made a public record 45 improperly, then the witness would have been harmed, for no good purpose. The District Court summarily denied the motion to unseal, [f]or reasons presented in the Government s response. We reject this argument made by the government, and therefore we reverse the District Court ruling on this issue.25 In short, our agreement with the government starts and ends with their observation that there exists a presumption of openness in all legal proceedings. United States v. Ochoa-Vasquez, 428 F.3d 1015, 1029 30 (11th Cir. 2005). But the government s argument minimizes what is at stake when that presumption of openness is overcome. First, the value of openness in criminal proceedings extends far beyond just the interests of any particular defendant. Rather, we have explained that: In balancing the public interest in accessing court documents against a party s interest in keeping the information confidential, courts consider, among other factors, whether allowing access would impair court functions or harm legitimate privacy interests, the degree of and likelihood of injury if made public, the reliability of the information, whether there will be an opportunity to respond to the information, whether the information concerns public officials or public concerns, and the availability of a less onerous alternative to sealing the documents. 25 This Court review[s] for abuse of discretion the refusal of a district court to unseal court documents. Romero v. Drummond Co., Inc., 480 F.3d 1234, 1242 (11th Cir. 2007). 46 Romero, 480 F.3d at 1246. Thus, while it is true that Dr. Jordan s privacy interests sit on one side of the balance, it is the interest of the public in accessing the information that rests on the other. Id. And, in this case, the public has a great interest in learning the contents of the Notice namely, learning the highly material fact that Dr. Jordan, a repeat government expert witness, abused his government authority and committed acts which could have been charged as felonies. To say that the defense would have preferred to use this information to discredit Dr. Jordan s testimony is almost certainly an understatement. Perhaps ironically, by arguing that there was no Brady violation in this case because the AUSA prosecuting Ignasiak was unaware of Dr. Jordan s history, it is actually the government that most persuasively highlights the value in unsealing the Notice. Indeed, should the Notice remain sealed, the significant likelihood is that in the next CSA prosecution in which Dr. Jordan testifies as an expert, both the prosecuting AUSA and the defense counsel will again be unaware of the highly relevant impeachment evidence contained in the Notice. And in that case, as in this one, should the truth ever come to light, the government could again point to its own ignorance and claim immunity from Brady error. Stated this way, we would have expected the government to condemn, rather than condone, such a problematic outcome. 47 But instead the government asserts that Dr. Jordan s privacy interest outweighs the public s right to know the extent of Dr. Jordan s involvement with the government. To be sure, in some cases a party may overcome the presumption of openness if it can show an overriding interest based on findings that closure is essential to preserve higher values and is narrowly tailored to serve that interest. Press-Enterprise Co. v. Superior Court of California, 464 U.S. 501, 510, 104 S. Ct. 819, 824 (1984). Indeed, the government correctly points to two categories of witnesses whose privacy interests are understandably paramount: victims in sex crime cases and criminal informants. Dr. Jordan is neither. Rather, he is an expert witness who, at a rate of $300/hour, voluntarily accepted employment which required him to testify against Ignasiak. Indeed, Dr. Jordan testified that he has been paid around $30,000 for his service as the government s expert in this and other cases. While the fact of his paid status does not make him amenable to any sort of unfair or immaterial character attack, it does greatly reduce, if not altogether eviscerate, his expectation to keep impeachment evidence private. The government is thus right that courts should protect witnesses like Dr. Jordan from unwarranted invasion into their privacy. But we cannot agree that impeachment evidence concerning a highly compensated and voluntarily appearing expert witness is either unwarranted or an invasion into that witness s privacy. 48 V. CONCLUSION For all the reasons above, we reverse Ignasiak s convictions and remand this case to the District Court for further proceedings. We further vacate the District Court s order sealing the government s notice filed under seal regarding Dr. Jordan. REVERSED, VACATED, and REMANDED. 49