OPINION

BERNICE BOUIE DONALD, Circuit Judge.

When a doctor first enters the practice of medicine, he or she swears to abide by a prime directive of the profession: "First, do no harm." Paul Volkman breached this sacrosanct tenet when he prescribed narcotics to addicts and individuals with physical, mental, and psychological frailties. A federal jury looked at Volkman's actions and found him guilty of breaking several laws, chief among them the law prohibiting the unlawful distribution of controlled substances. After receiving the jury's verdict, the district court sentenced Volkman to four consecutive terms of life imprisonment, to be served concurrently with a number of less-lengthy terms.

Volkman appealed his conviction and sentence and we affirmed the district court by published opinion. See United States v. Volkman, 736 F.3d 1013 (6th Cir.2013). The Supreme Court then granted Volkman a writ of certiorari and vacated our judgment. See Volkman v. United States, ___ U.S. ___, 135 S.Ct. 13, 190 L.Ed.2d 286 (2014). On remand, we are asked to consider, in light of Burrage v. United States, ___ U.S. ___, 134 S.Ct. 881, 187 L.Ed.2d 715 (2014), whether sufficient evidence of but-for causation supported Volkman's convictions under the Controlled Substances Act. Because we find the evidence of but-for causation sufficient and because Volkman's other allegations of error continue to lack merit, we AFFIRM the district court and resubmit this opinion, amended at Section IV.C, to address Burrage's but-for standard of causation.

I.

Paul Volkman is a former doctor who cast himself as a "pain management physician." Educated at the University of Chicago, Volkman holds an M.D. and Ph.D. in pharmacology from that institution. See Volkman v. United States Drug Enforcement Admin., 567 F.3d 215, 217 (6th Cir. 2009). Before the events leading up to his conviction, he was board-certified in emergency medicine and was a "diplomat" of the American Academy of Pain Management.

Despite his professional pedigree, Volkman fell into hard times in 2003. He had been sued on several occasions, settling some cases and losing others. Id. By the time his legal woes were over, he had no malpractice insurance and no job.

As part of his effort to rectify the latter, Volkman called Denise Huffman at the Tri-State Health Care clinic, asking about job opportunities. Eventually, Denise1 hired him to provide the clinic's medical services. They agreed that Volkman's salary would be $5,000 per week — eventually, the amount was upped to $5,500 per week.

Tri-State operated as a cash-only clinic. Pain medication was its bread and butter. At its peak, Tri-State and its staff saw an average of eighteen to twenty patients a day.

Volkman worked without incident during his first few months at the clinic. But approximately six months into the job, his practice encountered a major hiccup — local pharmacies refused to fill the clinic's prescriptions, citing concerns of improper dosing. Volkman's solution? Open a dispensary in the clinic. Volkman asked Denise's daughter, Alice Huffman Ball, to research the process for obtaining a license to operate a dispensary. Denise objected and raised concerns, but Volkman assured her that "he was a doctor[,] so he could dispense his own medication and he could take care of everything."

Volkman submitted to the Ohio Board of Pharmacy an application for a license to distribute controlled substances. Board representatives conducted an inspection of the clinic grounds, during the course of which they found a Glock in the safe where the drugs were stored. Despite this discovery, the Board issued a license after its initial inspection.

Agents from the Board conducted a follow-up inspection in December 2003. This time, they saw several problems with the new dispensary's practices. For instance, the dispensary logs were sloppily maintained; Volkman provided little oversight over recordkeeping processes. No licensed physician or pharmacist oversaw the actual dispensing process. Patients returned unmarked and intermixed medication.

By February 2004, the clinic took adequate measures to ameliorate the Board's administrative concerns. But the clinic still had its problems. Volkman was in charge of the dispensary, but did a poor job of regulating access — the drug safe's security was porous, with unauthorized personnel regularly accessing the pharmaceutical stockpile contained inside. Despite these issues, the dispensary saw much activity — it purchased 135,900 dosage units of oxycodone between July and December 2003, 457,100 dosage units for the entirety of 2004, and 414,200 dosage units between January and September 2005.

It eventually became clear that Volkman's medical practice followed a questionable pattern. Drug addicts, drug peddlers, or individuals otherwise not complaining of pain would come to see him as his "patients." Very little was done in terms of taking medical histories or conducting physical examinations. Volkman would regularly prescribe a drug cocktail consisting of opiates (such as oxycodone and hydrocone) as well as sedatives (diazepam, alprazolam, and carisoprodol; more commonly referred to as Valium, Xanax, and Soma). He had a tendency of first resorting to narcotics, disregarding first lines of treatment for pain management such as non-steroidal anti-inflammatory drugs (NSAIDs).

A federal investigation of Tri-State led to a search of the clinic facility on June 7, 2005. Medical personnel accompanying the investigative team saw that the clinic was in utter disarray. Urine specimen cups, filled with urine, were scattered all over the floor. The clinic had no equipment to view X-rays and MRI results. Miscellaneous pills were strewn all throughout the clinic premises.

Three months after the investigation, Denise terminated Volkman's employment because she "could no longer get along with him" and because there was "no control." In her words, "Dr. Volkman did what Dr. Volkman wanted to do." Volkman decided to open his own shop in Ohio — first in Portsmouth, and later in Chillicothe.

Twelve of Volkman's patients died during his tenure at Tri-State and during the early months of his new practice. Kristi Ross, Steve Hieneman, Bryan Brigner, and Earnest Ratcliff were four of these patients.

A grand jury returned an indictment against Volkman, Denise, and Alice, charging them with one count of conspiring to unlawfully distribute a controlled substance in violation of 21 U.S.C. § 841(a)(1) (the Controlled Substances Act or "CSA"), two counts of maintaining a drug-involved premises in violation of 21 U.S.C. § 856(a)(1), eight counts of unlawful distribution of a controlled substance leading to death in violation of 21 U.S.C. §§ 841(a)(1) and 841(b)(1)(C), and four counts of possession of a firearm in furtherance of a drug-trafficking crime in violation of 18 U.S.C. §§ 924(c)(1) and (2). Volkman was charged separately with five additional counts of unlawful distribution of a controlled substance leading to death, as well as two additional counts of maintaining a drug-involved premises. His co-defendants, Denise and Alice, entered into plea agreements with the Government and testified against their former colleague.

After a thirty-five-day trial, a jury convicted Volkman on the lone conspiracy count, seven counts of unlawful distribution that did not lead to death, four counts of unlawful distribution leading to death, four counts of maintaining a drug-involved premises, and one count of possessing a firearm in furtherance of a drug-trafficking offense. The jury acquitted him on one count of unlawful distribution, as well as one count of possessing a firearm in furtherance of a drug-trafficking offense.

The district court sentenced Volkman to four consecutive terms of life imprisonment for the counts of unlawful distribution leading to death, to be served concurrently with a sentence of 240 months for the counts of conspiracy and unlawful distribution not leading to death, 120 months for the drug-related premises counts, and 60 months for the firearm count, followed by three years of supervised release. Volkman timely appealed.

II.

Volkman divides his argument into four parts. He first argues that the district court erred by denying a proposed jury instruction derived from the Supreme Court's decision in Gonzales v. Oregon, 546 U.S. 243, 126 S.Ct. 904, 163 L.Ed.2d 748 (2006). Next, he contends that the Government's expert witnesses improperly provided legal conclusions as to whether Volkman's actions had a "legitimate medical purpose." Third, he claims that there was insufficient evidence to support the jury's guilty verdict on several charges. Finally, he challenges the reasonableness of his sentence on numerous grounds.

First, we turn to Volkman's jury-instruction argument. We review a denial of a proposed jury instruction for an abuse of discretion. United States v. Theunick, 651 F.3d 578, 589 (6th Cir.2011). Under this standard of review, we may reverse a district court's denial only if the proposed instruction "is (1) a correct statement of the law, (2) not substantially covered by the charge actually delivered to the jury, and (3) concerns a point so important in the trial that the failure to give it substantially impairs the defendant's defense." United States v. Franklin, 415 F.3d 537, 553 (6th Cir.2005) (citation and quotation marks omitted).

Volkman sought to have the following language included in the jury charge: "In other words, in order to find the defendant guilty, you must find that he used his prescription-writing power as a means to engage in the illicit drug-dealing and trafficking as conventionally understood." As we previously noted above, this language comes directly from the Supreme Court's decision in Gonzales. See 546 U.S. at 269-70, 126 S.Ct. 904. Accordingly, Volkman contends that the proposed language — copied verbatim from Gonzales — reflected a correct statement of the law that should have been included in the district court's instructions to the jury.

Verbatim, however, does not necessarily mean correct. Context is critical, and in the present context of federal criminal law, Gonzales provides us with little guidance.

We join our sister circuits in making this observation. In United States v. Lovern, 590 F.3d 1095 (10th Cir.2009), the Tenth Circuit commented on Gonzales' relevance — or lack thereof — in the setting of a criminal prosecution. The court explained that Gonzales dealt only with the question of the Attorney General's ability to define "legitimate medical purpose" in light of state medical standards to the contrary. See id. at 1100. Gonzales was decided in the setting of administrative law, not criminal law. In the court's view, Gonzales was not relevant to the criminal prosecution at issue in Lovern because there was "no interpretive rule seeking to define a practice as lacking any legitimate medical purpose, let alone a rule that conflict[ed] with a state's assessment of the legitimacy of that practice." Id. Instead, the Lovern court noted that the Government properly sought to prove the lack of a legitimate medical purpose by evidentiary means, leaving the question of what constitutes "the usual course of professional practice" for a jury to sort out. See id.

The Eighth Circuit made similar observations in United States v. Kanner, 603 F.3d 530 (8th Cir.2010), where it adopted and expanded upon the Tenth Circuit's reasoning. There, a defendant argued that the indictment should have included the same language from Gonzales that Volkman relies upon now. Id. at 533. The Kanner court rejected the defendant's argument by quoting the Lovern decision. Id. at 533-34. It added to the Lovern court's conclusions by noting that "Gonzales did not supplant the standard for violations of the CSA." Id. at 535. "Rather, post-Gonzales, knowingly distributing prescriptions outside the course of professional practice is a sufficient condition to convict a defendant under the criminal statutes relating to controlled substances." Id. (quoting United States v. Armstrong, 550 F.3d 382, 397 (5th Cir.2008), cert. denied, 558 U.S. 829, 130 S.Ct. 54, 175 L.Ed.2d 44 (2009)).

As our sister circuits have suggested, Gonzales did not impose new requirements to prove a violation of the CSA. Instead, the statement that Volkman now quotes was merely part of the Court's commentary about statutory intent, federalism, and rulemaking authority — none of which is at issue here. See Gonzales, 546 U.S. at 270, 126 S.Ct. 904.

In the past, we have endorsed a broad approach to determining what conduct falls outside the accepted bounds of professional practice so as to constitute a CSA violation, eschewing a preestablished list of prohibited acts in favor of a case-by-case approach. See United States v. Kirk, 584 F.2d 773, 784 (6th Cir.1978). Simply put, Volkman's proposed instruction would have needlessly narrowed the scope of the jury's inquiry to a question of whether Volkman engaged in "conventional" drug dealing and trafficking by using his prescription power. By narrowing the scope of the jury's deliberation in such a manner, Volkman's instruction would have been inconsistent with our endorsement of the broad approach, improperly cabining a decision that is properly left to the jury.

Moreover, the district court raised fair questions about the nebulousness of the proposed instruction. What does it mean for drug dealing and trafficking to be "conventionally understood"? If Volkman's goal was to conjure up the unsavory specter of "street" drug dealing — complete with imagery of shady characters conducting quick, suspicious handoffs — then his instruction was not an accurate statement of the law, for "street" drug dealing is not necessary to prove a violation of the CSA. It is not difficult to see, however, how a lay juror might think that proof of such "street dealing" was necessary for conviction under Volkman's proposed standard. In that respect, we agree with the district court that the instruction would have "muddied the waters," thereby providing an inaccurate statement of the law.

Gonzales did nothing to alter the reality that "`knowingly distributing prescriptions outside the course of professional practice is a sufficient condition to convict a defendant under the criminal statutes relating to controlled substances.'" Kanner, 603 F.3d at 535 (quoting Armstrong, 550 F.3d at 397). Volkman's proposed jury instruction improperly would have cabined the scope of what the jury could consider, thereby providing an inaccurate statement of law. Therefore, the district court correctly denied Volkman's request.

Our decision to uphold the district court's denial is reinforced by the second prong of our review: whether the jury charge issued by the district court substantially covered what Volkman sought to convey by his proposed instruction. We conclude that the jury charge was not only adequate, but an example of model instructions for cases such as this one.

After reviewing the elements of a § 841(a)(1) offense and the meaning of certain terms such as "distribute," "dispense," and "practitioner," the district court provided a lay explanation of the contours of the crime. It stated:

(DE 482, PageID 8588-89.) Next, the court elaborated on the burden of proof:

(DE 482, PageID 8589.) After explaining the textbook definition of "usual course of professional practice," the trial court connected the definition to the instant case:

(DE 482, PageID 8589-90 (emphasis added).) The court capped off its explanation with a discussion about the standard of care:

(DE 482, PageID 8590-91.)

We conclude that these instructions amply and accurately conveyed the meaning of "legitimate medical purpose" to the jury — the ultimate purpose of Volkman's proposed instruction. Not only were these instructions adequate to allay Volkman's concerns, they served as a model of clarity and comprehensiveness in defining the unlawful-distribution offense for a case involving a so-called "pill mill" doctor. The district court's instructions appropriately defined the contours of the offense without unduly cabining the jury's ability to consider a broad swath of evidence in determining whether Volkman's conduct had no legitimate medical purpose. See United States v. August, 984 F.2d 705, 713 (6th Cir.1992) ("There are no specific guidelines concerning what is required to support a conclusion that an accused acted outside the usual course of professional practice."). Hence, the district court properly rejected Volkman's proposed instruction.

III.

Next, we address Volkman's argument that the district court improperly admitted expert testimony containing a legal conclusion. We review a district court's admission of expert testimony for an abuse of discretion. In re Scrap Metal Antitrust Litig., 527 F.3d 517, 528 (6th Cir.2008). The evidentiary admission must be upheld unless the district court "base[d] its ruling on an erroneous view of the law or a clearly erroneous assessment of the evidence." Best v. Lowe's Home Ctrs., Inc., 563 F.3d 171, 176 (6th Cir.2009) (quotation omitted).

We accord district courts a "wide, but not unlimited, degree of discretion in admitting or excluding testimony [that] arguably contains a legal conclusion." United States v. Nixon, 694 F.3d 623, 631 (6th Cir.2012) (quoting Torres v. Cnty. of Oakland, 758 F.2d 147, 150 (6th Cir.1985)) (internal quotation marks omitted). A witness' testimony contains a legal conclusion only if "the terms used by the witness have a separate, distinct and specialized meaning in the law different from that present in the vernacular." Torres, 758 F.2d at 151. An expert may not opine on the overarching question of guilt or innocence, but he or she may "stat[e] opinions that suggest the answer to the ultimate issue or that give the jury all the information from which it can draw inferences as to the ultimate issue." Berry v. City of Detroit, 25 F.3d 1342, 1353 (6th Cir.1994).

Here, the controverted expert testimony followed a pattern. First, the Government would ask its expert witnesses about Volkman's history with a particular patient. After some back-and-forth about a patient's condition and the prescriptions Volkman dispensed, the Government would ask the expert whether he or she had an opinion as to whether the prescriptions fell within the scope of legitimate medical practice. The answer was typically no. Consider, for example, this exchange between the prosecution and Dr. Douglas Kennedy:

Volkman asserts that this type of testimony constituted an improper legal conclusion.

We disagree, and join at least two of our sister circuits in doing so. The Seventh Circuit's decision in United States v. Chube II, 538 F.3d 693 (7th Cir.2008), is particularly persuasive. There, two doctors were convicted of unlawfully distributing controlled substances. The Government introduced testimony by experts who concluded that the doctors' prescription practices did not conform to the "usual standards of medical practice" and were devoid of a "legitimate medical purpose." Id. at 697. A typical colloquy between the Government and an expert went something like this:

Id. at 698. The court explained that "what the jury heard was ... an opinion from the expert that no legitimate medical purpose existed for the prescription in question." Id. (emphasis added). It concluded that the district court did not abuse its discretion in allowing such testimony. Id. at 699. The Seventh Circuit was motivated by a practical consideration: it observed that "it is impossible sensibly to discuss the question whether a physician was acting outside the usual course of professional practice and without a legitimate medical purpose without mentioning the usual standard of care." Id. at 698.

In United States v. Schneider, 704 F.3d 1287 (10th Cir.2013), the Tenth Circuit arrived at a similar conclusion. That case involved two doctors who were convicted of unlawful drug distribution and health care fraud. The Government relied on experts who testified that one of the defendants "prescribed controlled substances `for other than legitimate medical purposes.'" Id. at 1294. Although the defendants challenged the admission of this testimony, their arguments were unavailing. After observing that an "ultimate-issue" concern only arises when "an expert uses a specialized legal term and usurps the jury's function," the Tenth Circuit concluded that the experts' use of the phrase "other than legitimate medical purposes" posed no issue. Id.

Much like the Government's experts in Chube II, the experts in this case merely provided opinions suggesting that Volkman had no legitimate medical purpose for issuing a particular prescription to a specific patient. See Chube II, 538 F.3d at 699. Certainly, there is the legal question of whether a prescription had a "legitimate medical purpose," but the question is hardly answered in isolation. Rather, the "lay" or, as we have previously described it, "vernacular" understanding of the phrase — i.e., the phrase as used in medical parlance — naturally informs the legal question. See id. at 698. Therefore, the legal understanding of the phrase "legitimate medical purpose" does not carry with it a "separate, distinct and specialized meaning" from its medical counterpart; instead, one elucidates the other. See Torres, 758 F.2d at 151; see also Schneider, 704 F.3d at 1294. We discern no error in the district court's admission of the expert colloquies that ended with the conclusion that certain prescriptions had no "legitimate medical purpose."

Two exchanges, however, are worthy of separate discussion. Dr. Steven Severyn and an area pharmacist, Mark Carroll, provided a medical-purpose conclusion on the stand, despite not having been asked about a particular individual's prescriptions. But even then, the questions were appropriately limited to the standard of care and the evaluation of certain drug combinations or drug quantities. Nothing in the record suggests that the two experts sought to usurp the jury's function by drawing a legal conclusion; instead, both experts applied their understanding of the standard-of-care to a limited sample of facts. Accordingly, no error lies in the admission of their testimony.

IV.

At the heart of Volkman's appeal lies six sufficiency-of-the-evidence challenges. He contends that there was insufficient evidence for the jury to convict him on the conspiracy and firearm charges; in addition, he argues that there was insufficient evidence to convict him of the drug-induced deaths of Kristi Ross, Steven Craig Hieneman, Bryan Brigner, and Earnest Ratcliff.

In reviewing these convictions, we ask whether, "after construing the evidence in favor of the Government, no rational trier of fact could have found the essential elements of a crime beyond a reasonable doubt." United States v. Ross, 703 F.3d 856, 882 (6th Cir.2012). The standard is a "steep climb," id. (quotation omitted), and circumstantial evidence alone can defeat a sufficiency challenge, United States v. Washington, 702 F.3d 886, 891 (6th Cir.2012).

A.

We start with Volkman's conviction under 21 U.S.C. § 846 — the conspiracy charge. To satisfy the statute's requirements, "the government must prove the existence of an agreement to violate the drug laws and that each conspirator knew of, intended to join, and participated in the conspiracy." United States v. Conrad, 507 F.3d 424, 432 (6th Cir.2007) (quotation omitted). "The connection between the defendant and the conspiracy need only be slight," United States v. Craft, 495 F.3d 259, 265 (6th Cir.2007) (quoting United States v. Crayton, 357 F.3d 560, 573 (6th Cir.2004)), and a "conspiracy may be inferred from circumstantial evidence which may reasonably be interpreted as participation in a common plan," Conrad, 507 F.3d at 432 (quotation omitted). In addition, "[a] tacit or material understanding among the parties to a conspiracy is sufficient to establish the agreement." Id. (quotation omitted).

Volkman claims that he had no knowledge of the conspiracy's "main purpose" — "to make as much money as possible by distributing and dispensing controlled substances." Instead, he asserts that he was hired with the sole understanding that he would be serving as a pain management physician. The clinic was, in the ex-doctor's view, a legitimate medical operation — complete with in-house and hospital drug screens, pill counts, and in-clinic monitoring of patients. According to Volkman, an illicit, drug-based profit motive did not fit into the big picture.

His recollection of events, however, turns a blind eye to quite a bit of evidence — evidence that supports his conspiracy conviction. The conspiratorial relationship began when Denise hired Volkman and paid him $5,500 per week in exchange for seeing those who wanted pain medication. There was evidence to suggest that, by the time area physicians began refusing to fill Volkman's prescriptions, the purported members of the conspiracy — Denise, Alice, and Volkman — were aware of the reality that the prescriptions from their clinic had no legitimate medical purpose. Instead of rectifying the pharmacists' issues with Volkman's prescriptions, the trio exacerbated the problem by continuing to issue prescriptions, cutting out the middleman, and opening their own dispensary. This, by itself, was enough for a jury to find that the trio executed a plan to unlawfully distribute controlled substances with no legitimate medical purpose.

And contrary to his assertions, there was sufficient evidence to show that Volkman was hardly the blissfully-ignorant doctor he now makes himself out to be. A rational trier of fact could have easily concluded that the entire enterprise of dispensing pills straight from the clinic was Volkman's brainchild. That same trier of fact could have determined it was highly unlikely that Volkman — a man who prided himself on knowing the inner workings of his clinic — was unaware of the clinic's profits. All this, combined with Volkman's role in rubber-stamping the distribution of prescriptions, could have easily persuaded a jury that Volkman knew of and actively participated in the charged conspiracy.

B.

The same holds true for Volkman's firearm conviction. He argues that there was insufficient evidence to convict him of possessing a firearm in furtherance of a drug-trafficking crime because (1) the Government did not establish possession; and (2) it did not prove that the weapon was used in furtherance of a drug-trafficking offense. We disagree.

First, we address the question of possession. For purposes of the firearms statute, "[c]onstructive possession is established when a defendant `knowingly has the power and the intention at a given time to exercise dominion and control over an object, either directly or through others.'" United States v. Kelsor, 665 F.3d 684, 692 (6th Cir.2011) (quoting United States v. Hadley, 431 F.3d 484, 507 (6th Cir.2005)). Here, there was evidence showing that Alice's father placed the gun in the safe containing the drugs "for security purposes." Volkman had access to the safe. The record also reveals that he had concerns about personal security, giving him a reason to use the gun should the need arise. Video evidence confirms that the Glock was in the safe in 2005; Volkman was still working at the clinic at the time. This was sufficient for a rational trier of fact to conclude that Volkman constructively possessed the firearm for which he was indicted and convicted.

But that does not end our inquiry, for Volkman argues that even if he constructively possessed the firearm, the fact that it was in a locked safe meant that the weapon was not possessed in furtherance of a drug crime, given its immediate inaccessibility. We, however, have concluded otherwise in the past. When a weapon is found in a locked safe placed alongside contraband, there is sufficient evidence for a jury to determine that a defendant is in possession of a firearm in furtherance of a drug-trafficking crime. See United States v. Mendizabal, 214 Fed.Appx. 496, 501 (6th Cir.2006); see also United States v. Cobbs, 233 Fed.Appx. 524, 535-36 (6th Cir.2007) (concluding that there was sufficient evidence for a jury to find that a locked safe with four handguns and crack cocaine nearby the safe constituted possession of a firearm in furtherance of a drug offense). Hence, Volkman's challenge to his firearm conviction must fail.

C.

The most serious of Volkman's convictions are the ones for unlawful distribution of a controlled substance leading to death. Volkman contends that a rational trier of fact could not conclude that his prescription practices resulted in the deaths of Kristi Ross, Steve Hieneman, Bryan Brigner, and Earnest Ratcliff.

A violation of the CSA occurs when a physician dispenses or distributes a controlled substance in a manner that is not authorized by law — i.e., the prescription is issued without "a legitimate medical purpose by an individual practitioner acting in the usual course of his professional practice." See 21 U.S.C. § 841(a)(1); 21 C.F.R. § 1306.04(a); see also Kirk, 584 F.2d at 784. To determine whether a physician has violated the CSA, the jury may undertake a "case by case analysis of evidence to determine whether a reasonable inference of guilt may be drawn from specific facts." Kirk, 584 F.2d at 784. When a physician violates the CSA in a manner that leads to the death of a patient, there is a mandatory sentence of twenty years to life in prison. See 21 U.S.C. § 841(b)(1)(C).

The Supreme Court held in Burrage v. United States that, "at least where use the drug distributed by the defendant is not an independently sufficient cause of the victim's death or serious bodily injury, a defendant cannot be liable for the penalty enhancement" under § 841(b)(1)(C). 134 S.Ct. at 892. In other words, use of the drug must have been a but-for cause of the victim's death or injury. Id. But-for causation exists where use of the controlled substance "combines with other factors to produce" death, and death would not have occurred "without the incremental effect" of the controlled substance. Id. at 888.

In the present case, Volkman concedes that the district court properly gave the jury a "but-for" causation instruction."2 Specifically, the court instructed the jury:

(DE 482, PageID 8593-94.) The district court's instruction notwithstanding, Volkman contends that no rational trier of fact court have found, as the jury did here, that death would not have occurred but-for the use of the oxycodone prescribed. We disagree. With respect to the deaths of these four individuals, there was sufficient evidence for a jury to conclude that (1) Volkman issued a prescription; (2) that had no legitimate medical purpose; (3) which was the but-for cause the victim's death. Thus, the Government satisfied its burden of proof under the CSA as interpreted by the Supreme Court in Burrage. Although Volkman attempts to cherry-pick evidence to dispute some aspect of each conviction, we are unpersuaded by his arguments.

1.

First, we have Kristi Ross. On March 8, 2004, Ross — an obese, 39-year-old woman — came into Volkman's office, complaining of lower back pain, cervical myalgia, and hypertension. Her pending divorce and her husband's efforts to take custody of their daughter compounded her stress. Volkman prescribed a cocktail of Soma (carisoprodol), Lorcet (hydrocodone), oxycodone, and Xanax (alprazolam). On this visit, and on Ross' prior visit on January 6, 2004, Volkman had increased her dosage of oxycodone while keeping constant her prescriptions for hydrocodone, carisoprodol, and Xanax. (DE 296, PageID 3426-27.)

Ross, however, suffered from a number of risk factors. In addition to her hypertension and obesity, she potentially suffered from sleep apnea, as well as some form of lung disorder. In other words, she was already at risk of breathing difficulties, and Volkman's drug cocktail exposed her to a greater risk of death. A rational trier of fact, relying on expert testimony, could have found that there was no legitimate medical purpose for this combination of drugs.

On March 9, 2004 — one day after her last visit to Volkman — Ross was found dead, with the very drugs prescribed to her in her purse. A toxicology report revealed that, at the time of her death, Ross had benzodiazepines and opiates in her system — a toxic combination "consistent with [Volkman's] prescriptions." (Id. PageID 3429.) Dr. Michael Policastro, a board certified emergency room physician and toxicologist, concluded that Ross died of multi-drug death. (Id.) The bottle of oxycodone in Ross' purse contained 26 oxycodone pills; the prescription, filled the day before, had been for 90 pills. (DE 448, PageID 6497-98.) This evidence of the increase in Ross' prescription for oxycodone just one day before her death — coupled with evidence that the prescriptions for the other drugs found in her system remained constant — was sufficient for a rational trier of fact to conclude that Volkman's unlawful prescription of oxycodone was a but-for cause of Ross' death. Burrage, 134 S.Ct. at 892.

Moreover, expert testimony corroborated the inherent risks of Volkman increasing Ross' dosage of oxycodone. Dr. Policastro testified that doubling the daily dose of oxycodone, as Volkman did here at Ross' final visit, accelerates the risk that there will be a "[d]ecline of your breathing." (DE, PageID 3427.) Dr. Policastro further testified that while "your body can... become[] tolerant to that breathing problem[,] ... [a]s time progresses and doses of pain medications increase, there is a point at which you cannot compensate for that." (Id. PageID 3359.) Likewise, Dr. Kennedy testified that Volkman's final oxycodone prescription to Ross was "scary" and "extremely dangerous" if taken as prescribed, particularly when used in combination with other drugs that also suppress breathing. (DE 301, PageID 3781-82.)

Volkman takes issue with the accuracy of the post-mortem procedures used on Ross' body — specifically, he argues that fluids should have been extracted from Ross' femoral artery, as opposed to her eye. While this argument may be perfectly valid, we are not in a position to entertain it; it is for the jury to decide which evidence — scientific or otherwise — to credit in making its determination. United States v. Washington, 715 F.3d 975, 979 (6th Cir.2013) (citing Jackson v. Virginia, 443 U.S. 307, 319, 99 S.Ct. 2781, 61 L.Ed.2d 560 (1979)). It was well within the jury's ambit for it to find the methodology behind Ross' forensic examination reliable.

Volkman also advances the related argument that the concentration of oxycodone extracted from Ross' eye was below the level normally associated with death resulting from oxycodone. This argument, however, is not inconsistent with Dr. Policastro's testimony. Dr. Policastro specifically testified that despite the fact that Ross' "oxycodone level was slightly below the range of levels associated with death," death could nevertheless result due to the "accelerating pattern of the dosages of the meds prior to death." (DE 296, PageID 3429.) This testimony, coupled with other testimony regarding the way in which oxycodone interacted with the other drugs and Ross' existing health risks (including her preexisting lung problems), was sufficient for the jury to conclude that Volkman's unlawful prescriptions resulted in Ross' death, thus justifying his conviction under § 841(b)(1)(C).

2.

Next, we have Steven Craig Hieneman. Hieneman was a 33-year-old man with cardiovascular problems. Volkman had seen Hieneman as a patient for quite some time-the doctor had prescribed oxycodone in the past, mostly in five-milligram doses. Eventually, these doses of oxycodone were ramped up to thirty-milligram doses. Hieneman saw Volkman on February 22, 2005, complaining of pain in his wrist, neck, lower back, and shoulder. Hieneman reported that, without medication, he was sleep-deprived, convulsed involuntarily in his hand and shoulder, and his pain was above average.

That day, Volkman prescribed Disalcid, Valium (diazepam), and Percocet (oxycodone) to treat Hieneman's ailments. On April 19, 2005, Volkman changed Hieneman's prescriptions; he issued prescriptions for Xanax (alprazolam), oxycodone, and Valium (diazepam). The prescriptions, particularly the one for oxycodone, were disturbing in light of the fact that Hieneman had a past addiction to Oxycontin and was battling other forms of substance abuse.

Volkman knew about Hieneman's history, yet issued prescriptions for oxycodone. In addition, the doctor knew he was prescribing these drugs to a man with severe hypertension and heart problems associated with drug use. Given these dangers, a rational trier of fact could have concluded — as one of the Government's experts did — that Volkman's prescription practices with respect to Hieneman were not within the scope of legitimate medical practice.

In addition, there was sufficient evidence for a rational jury to find but-for causation between Hieneman's prescription for oxycodone and his death. Hieneman died only twelve hours after receiving his prescriptions from Volkman. A deputy coroner concluded that Hieneman died an opiate drug-induced death. (DE 287, PageID 2800.) Of the drugs in Hieneman's system, oxycodone was the only opiate. The signs were all there: Hieneman's right cheek was blanched, his airways were partly blocked, his eyes were hemorrhaged under the conjunctivae, and he suffered from pulmonary edema. The link to Volkman? Bottles of oxycodone, diazepam, and alprazolam were found in Hieneman's home — i.e., the very drugs Volkman prescribed to Hieneman the day before. Dr. Christian Rolf, a medical examiner and board certified pathologist who performed Hieneman's autopsy, testified the oxycodone in Hieneman's blood was "high enough [to constitute] a toxic level" — even for a person who had built up a tolerance to oxycodone. (DE 287, PageID 2801, 2804.) Likewise, Dr. Policastro testified that the level of oxycodone in Hieneman's blood after his death was "within the range of levels associated with deaths." (DE 296, PageID 33443.)

Volkman cites evidence identifying Hieneman's cause of death as the "combined effects" of oxycodone, diazepam, and alprazolam. The Government was not required to prove, however, that oxycodone was Hieneman's only cause of death. On the contrary, but-for causation exists where a particular controlled substance — here, oxycodone — "combines with other factors" — here, inter alia, diazepam and alprazolmam — to result in death. Burrage, 134 S.Ct. at 888. The Government presented sufficient oxycodone-specific evidence for a rational jury to find that, "without the incremental effect" of the oxycodone, Hieneman would not have died. Id.

Volkman further argues that these very prescription bottles undermine the case against him: the issuing doctor for one bottle of oxycodone was identified as a Dr. Scott Lance, not Volkman. But there was sufficient evidence for the jury to discredit the theory that it was Lance, and not Volkman, who was ultimately responsible for Hieneman's death.

First, Volkman's patient notes acknowledged the existence of a Dr. Lance as a prescribing physician for Hieneman, but remarked that Lance had prescribed only Valium (diazepam). Relatedly, the notes indicated that Hieneman no longer had any of the medication Lance prescribed for him; it was gone at least a month prior to Hieneman's February 2005 visit to Volkman's clinic. Second, while one bottle of oxycodone was labeled as having been prescribed by Dr. Lance, the other bottles found at the scene did not have a prescribing physician listed. Finally, there was a close temporal nexus between Volkman's issuing of prescriptions and Hieneman's death. The relevance of this nexus is strengthened by the fact that Hieneman was found dead with the very drugs Volkman prescribed. Therefore, we conclude that the Government provided sufficient causal evidence to link Volkman to Hieneman's death.

We are also unpersuaded by Volkman's argument that Hieneman's underlying heart condition could have caused his death, irrespective of the drugs he consumed. The jury was free to reject this theory by crediting the testimony of one of the Government's experts, who concluded that "there was no notation of clot, no notation of infarct, scar, plaque, hemorrhage or anything like that," thereby suggesting that coronary disease was not the primary cause of Hieneman's death. Again, the existence of other potential contributing causes of death is irrelevant so long as the Government presented sufficient evidence that oxycodone was a but-for cause of Hieneman's death. Burrage, 134 S.Ct. at 888.

3.

Following Hieneman, we have Bryan Brigner. On August 4, 2005, Brigner came to see Volkman for treatment of his lower back pain. Brigner was disabled and had been on disability benefits for several years. When Volkman spoke with Brigner, Brigner denied having any other health problems. In truth, however, Brigner suffered from hypertension and arteriosclerotic cardiovascular disease.

When the two met that August day, Brigner was already on a regimen of Lortab (hydrocodone), oxycodone, Valium (diazepam), and Soma (carisoprodol). The oxycodone made him sick, so Volkman tried a smaller dose — but not for long. On September 30, 2005, Volkman prescribed a similar cocktail of Soma (carisoprodol), Valium (diazepam), Disalcid (a non-steroidal anti-inflammatory drug), Lortab (hydrocodone), and a strong dose of oxycodone. The oxycodone prescription was for the same dose that made Brigner sick prior to his August visit.

The new cocktail made little sense, especially given Brigner's underlying cardiac condition — a condition that weakened his ability to resist opiate toxicity. Volkman's notes on Brigner were also unhelpful; as one expert noted, it appeared as if there was "a lot of stuff ... but really the basics [weren't] covered in a useful, functional, medical fashion, just a bunch of stuff that's thrown up there." A rational trier of fact could have taken these facts and concluded that Volkman's prescriptions for Brigner did not bear the hallmarks of having a legitimate medical purpose.

As with Ross and Hieneman, Brigner was found dead shortly after his last visit with Volkman. Brigner died within just forty-eight hours of this final visit. Two causes of death were listed on the coroner's report: acute multiple drug intoxication and cardiopulmonary arrest. Brigner's hypertension and arteriosclerotic cardiovascular disease were reported as contributing factors. A toxicology analysis revealed that Brigner had taken oxycodone, hydrocodone, diazepam, haloperidol, and Sertraline just prior to his death. The first three — the drugs prescribed by Volkman — were enough to kill a person by respiratory depression. Put differently, there was enough for a rational jury to conclude that Volkman's drug cocktail led to Brigner's death.

Despite this evidence, Volkman attacks his conviction on three evidentiary grounds. First, he claims that if Brigner had taken Volkman's prescribed medication as instructed, he would not have overdosed because he would have built up a tolerance. Second, he asserts that Sertraline and haloperidol — two drugs that Volkman did not prescribe but which were found in Brigner's system — could have been responsible for Brigner's death. Third, he argues that underlying heart conditions, such as cardiomegaly (an enlarged heart) or unstable plaque, could have triggered the cardiac incident leading to Brigner's death.

For each one of these assertions, there is an evidentiary counterweight. First, a rational jury could have credited Dr. Bryan Castro's testimony that, of "[t]he medications ... included with [Brigner], none of them would give him opiate tolerance." Second, while the evidence did suggest that Sertraline and haloperidol can be lethal, they are only toxic in high doses not present here. The trier of fact could have concluded that the levels in Brigner's bloodstream were too low to support such a theory. By contrast, testimony from both Dr. Castro and Dr. Kennedy established that the level of oxycodone present in Brigner's blood was toxic or potentially toxic. (DE 315, PageID 4373; DE 300, PageID 3647-48, 3650.) Further, as with Ross, testimony established that Brigner's final prescription for oxycodone represented a significant increase relative to his previous prescriptions. Dr. Kennedy testified that, at this final visit, Volkman increased Brigner's prescription for oxycodone from 30 mg a day to 240 mg a day. (DE 315, PageID 4369.) On the same visit, Volkman kept constant his prescriptions for Valium and Soma and increased only slightly the prescription for hydrocodone, from 60 mg to 80 mg. (Id.)

Third, given the fact that there were no signs of a myocardial infarction — the usual indicator of complications arising from arteriosclerotic cardiovascular disease — a rational trier of fact was not compelled to accept Volkman's argument that an underlying condition led to Brigner's death.

Hence, we affirm this conviction.

4.

Finally, we have Earnest Ratcliff. Ratcliff was a 38-year-old man suffering from pain in his back, ankles, and feet. He had a history of drug addiction and minor drug dealing.

Volkman saw Ratcliff on October 21, 2005. The doctor issued prescriptions for 240 thirty-milligram doses of oxycodone, 90 doses of Soma (carisoprodol), 240 doses of Lortab (hydrocodone), and 90 doses of Xanax (alprazolam) — 660 pills in all. Given the sheer drug quantity alone, a rational trier of fact could have concluded that there was no legitimate medical purpose to the prescriptions, given the risk that Ratcliff would not comply with drug protocol and the attendant risk of an adverse outcome.

The next day, Ratcliff and his wife Melissa went to pick up the drugs from a drugstore in Columbus. After obtaining the drugs, he proceeded to consume — or, more exactly, snort — the pills on the way home. Ratcliff was lethargic and asleep for most of the drive. His wife found his behavior unusual, in light of his past history with the drugs.

Mrs. Ratcliff found her husband dead the next morning — just two days after his visit to Volkman's office. No autopsy was performed, but a toxicology report revealed that oxycodone, methadone, hydrocodone, diazepam, and nordiazepam were in his bloodstream. From this, a rational trier of fact had a sufficient evidentiary basis for concluding that Ratcliff's death-caused by "multi-drug intoxication" — was Volkman's fault.

Volkman argues that the methadone alone could have been lethal. Methadone, of course, was the one drug that he did not prescribe. Indeed, he points out that one of the Government's own experts testified that the methadone was "very significant" in Ratcliff's death.

There was sufficient evidence, however, for a jury to disregard the possibility that methadone acted as the lone lethal agent. First, it was a cumulative effect of toxicity, not a single drug, that was responsible for Ratcliff's death. Second, Ratcliff took methadone earlier in the week — prior to taking the drugs that Volkman gave him — and was functional after consuming the methadone. A jury could have used this as the basis for excluding the possibility that methadone alone was responsible for Ratcliff's death. Finally, a rational trier of fact could have credited the testimony of the toxicologist who examined Ratcliff's blood and agreed with her conclusion that oxycodone, not methadone, was present in the "most significant level" causing Ratcliff's death. (DE 300, PageID 3642.) Accordingly, we conclude that there was sufficient evidence to support Volkman's conviction for unlawful distribution resulting in the death of Earnest Ratcliff.

V.

The final salvo of Volkman's appeal attacks the reasonableness of his sentence. "This Court reviews criminal sentences for both substantive and procedural reasonableness," and uses the "deferential abuse-of-discretion standard" to determine the propriety of the sentence. United States v. Stubblefield, 682 F.3d 502, 510 (6th Cir.2012) (citing Gall v. United States, 552 U.S. 38, 51, 128 S.Ct. 586, 169 L.Ed.2d 445 (2007); United States v. Battaglia, 624 F.3d 348, 350 (6th Cir.2010)).

Two of Volkman's sentencing arguments focus on the district court's application of certain Guidelines enhancements. Volkman also disputes the district court's drug-quantity determination. We construe these contentions as procedural reasonableness challenges. See Stubblefield, 682 F.3d at 510; see also United States v. Daniels, 506 Fed.Appx. 399, 399 (6th Cir. 2012) (treating a drug-quantity objection as a procedural-reasonableness issue). As for Volkman's sentencing-disparity argument, we view that as a substantive reasonableness challenge. See United States v. French, 505 Fed.Appx. 478, 479-80 (6th Cir.2012).

A.

Volkman first takes issue with the district court's application of the vulnerable-victim enhancement. He asserts that the district court improperly based the enhancement on his victims' status as drug addicts.

Section 3A1.1(b) of the Sentencing Guidelines provides for a two-level enhancement in instances where "the defendant knew or should have known that a victim of the offense was a vulnerable victim." The Application Notes, however, state that the victim must be "an unusually vulnerable victim." U.S. Sentencing Guidelines Manual § 3A1.1(b) cmt. n. 2 (emphasis added).

In the past, we have acknowledged a tension between the Guideline language and the Application Notes. See United States v. Lukasik, 250 Fed.Appx. 135, 138 (6th Cir.2007). We need not resolve that tension today, however, because Volkman's conduct fell within the parameters of both.

We agree with Volkman in concluding that drug addiction, standing alone, cannot serve as the basis for applying the enhancement in cases such as this one. Drug addicts are not necessarily vulnerable victims for purposes of section 3A1.1, and we decline to categorically classify them as such. See United States v. Amedeo, 370 F.3d 1305, 1317 n. 10 (11th Cir.2004).

But the circumstances of this case nevertheless support application of the enhancement. See id. (noting that application of the enhancement is "highly fact-specific and must take into account the totality of the circumstances"). It is true that the Pre-Sentence Investigation Report, as well as the Government's arguments at sentencing, focused heavily on the victims' drug addiction. If this had been the sole basis for the district court's decision to apply the enhancement, then reversal would be warranted.

The court, however, made additional findings regarding specific victims' ailments. It noted the mental and emotional frailties of some of Volkman's patients by stating that "[s]ome patients had serious psychiatric problems[,] [s]ome even had prior suicide attempts." The court made these observations based on statements that Volkman made during the sentencing proceedings. We will not overturn a district court's fact findings unless they are clearly erroneous. See United States v. Moon, 513 F.3d 527, 539-40 (6th Cir. 2008). These fact findings were not clearly erroneous. Under either the Guideline or its accompanying note, the findings were sufficient to justify application of the vulnerable-victim enhancement. See U.S. Sentencing Guidelines Manual § 3A1.1(b) cmt. n. 2 ("[V]ulnerable victim means a person ... who is unusually vulnerable due to ... mental condition[.]"). Therefore, we discern no abuse of discretion here.

B.

Because we conclude that application of the vulnerable-victim enhancement was proper, we must next address Volkman's argument that applying both that enhancement and the special-skill enhancement was impermissible double counting. Such double counting occurs when "precisely the same aspect of the defendant's conduct is factored into his sentence in two separate ways." United States v. Lay, 583 F.3d 436, 447 (6th Cir. 2009) (quoting United States v. Farrow, 198 F.3d 179, 193 (6th Cir.1999)) (quotation marks and modifications omitted). "We review a district court's legal conclusions regarding the Sentencing Guidelines de novo." Moon, 513 F.3d at 540 (citation omitted).

The two enhancements at issue focus on different aspects of the case. Courts apply the special-skill enhancement when a "defendant ... use[s] a special skill, in a manner that significantly facilitate[s] the commission or concealment of the offense." U.S. Sentencing Guidelines Manual § 3B1.3. A special skill is defined as "a skill not possessed by members of the general public and usually requiring substantial education, training or licensing." Id. § 3B1.3 cmt. n. 4. The Guideline notes specifically list doctors as one group of individuals possessing special skills. Id. In contrast — and as the district court properly noted — the vulnerable-victim enhancement focuses on the nature of the victim, as opposed to the offender. See id. § 3A1.1(b) cmt. n. 2.

The fact that a defendant is a doctor — and his victim a patient — is insufficient for applying the vulnerable-victim enhancement. See United States v. Stokes, 392 Fed.Appx. 362, 371 (6th Cir. 2010) ("[T]he traditional doctor-patient relationship, on its own, provides an insufficient basis for applying the vulnerable-victim enhancement."). It is, however, sufficient for applying the special-skill enhancement, especially in light of the fact that Volkman would not have been in a position to abuse his prescription power without his medical license. See 21 U.S.C. § 823(f) (detailing the CSA's registration requirements); see also id. § 802(21) (defining "practitioner"); United States v. Moore, 423 U.S. 122, 140 n. 17, 96 S.Ct. 335, 46 L.Ed.2d 333 (1975). Given the differences in how the enhancements apply, we discern no impermissible double-counting here.

C.

We can readily dispose of Volkman's remaining arguments. First, Volkman argues that the district court did not account for legitimate prescriptions in arriving at its drug-quantity determination. Even if we assume that the district court erred by concluding that Volkman's drug-quantity determination warranted a base offense level of 38, such an error would be harmless. Volkman's convictions for unlawful distribution leading to death, standing alone, warrant the same base offense level. See U.S. Sentencing Guidelines Manual § 2D1.1(a)(2) (specifying a base offense level of 38 "if the defendant is convicted [under § 841(b)(1)(C)] ... and the offense of conviction establishes that death or serious bodily injury resulted from the use of the substance"). Consequently, we could not reverse on this ground. See United States v. Tandon, 111 F.3d 482, 491 (6th Cir.1997) (holding that a tax-loss calculation error was harmless because the base offense level would be unaffected).

We are equally unpersuaded by Volkman's sentencing-disparity argument. Four life sentences, one for each § 841(b)(1)(C) conviction, may appear relatively disparate. Other doctors in a similar position may have been sentenced to a less lengthy term of imprisonment. But each of Volkman's life sentences fell within the applicable Guidelines ranges. When a sentence is within-Guidelines, § 3553(a)(6) is an improper vehicle for challenging that sentence. See United States v. Swafford, 639 F.3d 265, 270 (6th Cir.2011). Thus, we conclude that Volkman's sentence was substantively reasonable.

VI.

We AFFIRM Volkman's convictions and sentence.