Defending Federal Overdose Cases

A gem of historical fiction follows the life of Balthasar Russow as he composes his Chronicle of Old Livonia.

https://wepresent.wetransfer.com/stories/art-of-the-trial

A primer on distribution resulting in death charges for defendants and their attorneys.

The federal distribution resulting in death (DRD) statutes (21 U.S.C. § 841(a)(1), (b)(1)) have existed for half a century, but only recently have federal prosecutions for overdoses become widespread. Consequently, defending overdose cases may be new territory for many defense attorneys. Although I am not an attorney I have spent the last half decade fighting a DRD charge and have learned a lot in this time that may be helpful to similarly situated defendants and their lawyers (and I spent several years prior to my charge acquiring some first-hand experience with opioids and the drug scene). Ultimately, I did take a plea deal, but not without first planning for trial and considering carefully all the evidence and possible strategies in anticipation of going in that direction. This post is essentially a summary of that planning and speculation about potential strategies.

But-For Causation

Firstly, and most importantly, what is the government trying to prove? A lot of people, even attorneys, get this question wrong, so both defendants and their attorneys should read Burrage v. United States, 571 U.S. 204, 134 S.Ct. 881, 187 L.Ed.2d 715 (2014) carefully and then read it again several more times. The accusation is that you (or your client) knowingly distributed (not necessarily sold) a controlled substance to someone who died from the use of that substance and would not have died “but for” that distribution having occurred. Intent is not a required element, nor is foreseeability; the only men’s REA requirement is the intent element of the lesser included offense of distribution (because Congress apparently intended this to be a sentencing enhancement and did not anticipate Apprendi). Knowledge of the exact substance is not required, only knowledge that the substance being distributed was a controlled substance. In other words, the greater offense of causing death is a strict liability offense and most standard homicide defenses will not be available. There are a couple of defenses, REA however, that might work. Sometimes.

Multiple Drugs

Despite setting an extremely low standard for the resulting in death element, the Supreme Court actually overturned Burrage’s conviction (after which the district court gave most of the time back to him one probation violation at a time). A but-for cause can be as insignificant as “the straw that broke the camel’s back”. However, something is not a but-for cause if the supposed effect would have happened anyway, unless it would have been an independently sufficient cause. The deceased person in Burrage’s case had multiple drugs in his system and the government’s expert witnesses were not willing to testify that the other drugs would not have been enough to kill him absent Burrage’s heroin. However, if the quantity of heroin had been enough to be potentially lethal by itself then it would have been an independently sufficient cause and the conviction would have been upheld.

A point that may be overlooked by attorneys who are not drug users is that the same type of drug may have been procured from multiple sources, and so the metabolite levels in the toxicology report will not necessarily be indicative of the quantity of drugs provided by the defendant that were ingested. Consider this: If you have an open container of milk in your refrigerator do you wait until it is empty before buying a new container? After you buy the new container do you open it immediately or do you finish the old one first? Opioid users live in constant fear of their stash running out and will always try to reup before this happens. After reupping they may keep the sacks separate, or they may mix them together, or they may celebrate by finishing whatever is left of the old sack (which may be a lot) all at once. Is it possible to determine which situation actually occurred? In principle, yes. If the government was serious about determining guilt and not simply trying to obtain a conviction then they could (but probably didn’t) test the contents of the recovered baggies for cutting agents and compare the results to the toxicology and/or autopsy results, assuming they know and can prove which sack came from the defendant. Or maybe the sack attributed to the defendant still weighs exactly what he is alleged to have sold. The details will vary from case to case, but if the deceased person was acquiring drugs from multiple sources (I have picked up from two people in one day before, when I wasn’t expecting both to come through) then the mere presence of that drug’s metabolites in the toxicology does not prove that the defendant’s drugs were ingested or that they were a but-for cause of death.

Middlemen

The presence of a middleman will not necessarily break a causal chain. But it may weaken it. The set of proximate causes is almost a subset of the set of but-for causes, but a proximate cause might fail to be a but-for cause if the effect was inevitable. An effective defense attorney probably wouldn’t try to argue to a jury that “he would have bought it from someone else anyway”, but if there was a middleman who actually is a distributor and not just the deceased person’s friend who picked up for him then it actually is very possible that the middleman not only might have but actually was buying from multiple sources, and just as a bar will not stop selling alcohol just because a single brand of beer gets discontinued a professional drug seller also will likely not stop selling drugs if one of his suppliers stops selling to him. If it happens that the middleman remains free and continues to distribute drugs while the defendant is detained pending trial then serious reasonable doubt exists as to whether the alleged harm would not have occurred but for his distribution. That is the fundamental question, and when a middleman exists the secondary question is the extent to which this creates reasonable doubt as to what would have happened had the defendant not distributed any drugs.

Toxicology

As high as a person’s tolerance may have been, it will probably not be possible to convince a jury that someone who is indisputably dead did not die of what he almost certainly did die of. But if there exists a competing possible cause of death then tolerance can call into question the defendant’s drugs as the cause of death. Alcohol and benzodiazepines, for instance, are a notoriously deadly combination, and if a high tolerance creates reasonable doubt about the defendant’s drugs being an independently sufficient cause of death then the question as to whether or not benzos and alcohol alone would be independently sufficient becomes relevant. Most overdoses involve people with low tolerances, but the government will try to apply average metabolite concentrations to all overdose cases leaving it to the defense to rebut their exaggerated assertions. In my case the lead detective sent an email to the prosecution about how many people could have been killed by the gram of china recovered (assuming it to be 100% pure, of course), a gram that I could have smoked in a day when my tolerance was high. (Side Note: All defense attorneys should read How to Lie with Statistics by Darrell Huff.)

Probably you won’t be able to quantify the tolerance of the deceased person. Instead, your aim will be to establish the possibility that his tolerance was not low and shift the burden onto the government to prove that it was. Someone claiming to be dopesick or exhibiting symptoms of dopesickness has a tolerance. Someone with detectable amounts of metabolites in his hair almost certainly has a tolerance. Someone with drug metabolites in his urine may have a tolerance. Someone with metabolites only in his blood, or with mostly unmetabolized opioids in his blood, very likely does not have a tolerance. Why is that?

Heroin, codeine, oxycodone, hydrocodone, hydromorphone, and oxymorphone are all morphine derivatives. They have varying half-lives but roughly they need to be ingested every 8–12 hours to prevent withdrawals. After they are ingested they are metabolized and it is the metabolites that are detected in urine and hair follicle tests. Metabolization ceases when a person dies. heroin metabolites are detectable in urine for roughly 1–3 days. Consequently, a person with metabolites in his urine has used heroin recently, and prior to the dose that killed him because that dose would not have had time to metabolize into the urine. Fentanyl is not a morphine derivative and so the presence of morphine in the urine but not the blood suggests multiple drug sources. Conversely, fentanyl in the urine without morphine suggests that he was using, and thus had a tolerance to, fentanyl, and thus did not die because he ingested fentanyl when he was expecting heroin or oxycodone.

If there is a question as to the source of the drugs then comparing and contrasting the blood, urine, and hair can help to establish which drugs actually caused death. Does the blood match the urine? If the defendant was his regular supplier then he most likely, unfortunately, was a but-for cause of death (so maybe don’t go to trial). Otherwise, it appears that the buyer was still using his stash from the day prior and maybe never touched what he bought from the defendant. Does the blood not match the urine? You now have evidence that there were multiple sources which calls into question the defendant being a but-for cause of death, though it might be better to not pursue this if there is evidence that the defendant was selling fentanyl and the urine contains morphine sans fentanyl metabolites. If buprenorphine is present in the blood and/or urine then effective tolerance will be substantially higher because buprenorphine is a partial agonist that is able to block opioid receptors from the binding of full agonists like morphine and fentanyl (this is the same mechanism responsible for the overdose reversing effects of naloxone).

As for the buyer claiming to be dopesick, take that with a grain of salt. The government will use this to argue that the defendant is the only possible source because if the deceased person wasn’t out of drugs then he wouldn’t be dopesick (see, for instance, United States v. McKinnie, 21 F.4th 283 (4th cir. 2021)). Just as sellers will often say they are “almost there” when they haven’t even left yet, buyers will say they are dopesick when they still have a stash in an effort to entice sellers to hurry up.

Counterfeit Pills

Fentanyl has become a hobgoblin, which means part of the defense is going to be clearing up misconceptions and mitigating any prejudice that the government might attempt to impart on the defendant by playing up these misconceptions. Maybe the defendant was selling counterfeit “M30” pills with fentanyl in them, but that by itself does not prove that the buyer believed that they did not contain fentanyl (see my post Overdose or Poisoning), nor does it prove that he only ingested one pill, and it is imperative for the defense to put the government to their burden of proof and not let any unfounded assumptions go unopposed. The first time I did “china” I did not know that it wasn’t just powder heroin, I didn’t even know what fentanyl was (though I was cautious regardless, because I was doing an unfamiliar drug from an unfamiliar source), but I eventually came to learn that “china” was fentanyl (which even the detective on my case acknowledged to be accepted slang). Similarly, it now seems to be common knowledge that street oxies are fentanyl, and that street Xanax, also, is often cut with fentanyl (and it is not hard to tell the difference between clandestinely pressed pills and pharmacy pills). This may not help much at trial, but at sentencing the 4-level misrepresentation enhancement (U.S.S.G. § 2D1.1(b)(13)(A)) might be avoided if a convincing case can be made that “M30” is established slang for fentanyl pills and is not an attempt to represent fentanyl as pharmaceutical oxycodone.

Also important is the quantity of fentanyl actually in the pills. Deadliness is in the dose and an M30 containing only .5mg of fentanyl would not be anymore deadly than a pharmaceutical M30, and even less an M80 (on that note, be sure to preemptively oppose the government prejudicially referring to fentanyl as “deadly fentanyl”). Fentanyl is a Schedule II controlled substance which means it has, according to the federal government, accepted medical use, and thus it is possible for it to be dosed properly and used safely. If the pills were not dosed properly then most likely it was not the defendant who manufactured them which can be argued as a mitigating factor at sentencing.

Miscellaneous Evidence

Most likely it will be futile to argue that the defendant did not meet with the deceased person and distribute drugs to him. A jury will want an explanation for how he died and the defendant will be the easiest person to blame, but looking at all the evidence together may reveal alternative possibilities. One of the most important pieces of evidence is the deceased person’s phone. The government will provide you with the incriminating text messages to and from the defendant immediately, but they will likely not be as forthcoming with other conversations involving other potential drug sources (it took me four years and three lawyers to gain access to the entire contents of the cellphone). If specific amounts are discussed then it might be possible to estimate how long a sack purchased from someone else should have lasted and thus whether it had likely been exhausted when the incident offense occurred.

Paraphernalia should always be recoverable. Needles, syringes, used foil with charred opioid remains on it, empty baggies (to be scraped for residue when all other sources of drugs have been depleted), tooters, Suboxone strips. Even if a sack is recovered that can be attributed to the defendant, if it doesn’t match the residue in the needle/syringe/tooter/foil then it is possible that it hadn’t been touched yet (it might match the blood, though, because a user might test it when he picks up but then set the new sack aside until the old one is finished). Xanax and oxycodone pills could be potential alternate sources of fentanyl in the toxicology. If any paraphernalia was not recovered that should have been this could be brought up to the jury. I served on a jury once and I learned from that experience that juries do not like feeling like their time is being wasted by a plaintiff that does not take its own case seriously, where it seems like the government was hoping to win with a half-assed investigation and dimwitted jury.

Did the deceased person have a smart watch? One of the key questions will likely be whether or not he was going through withdrawal before he died. In the first week or so of withdrawal, my pulse would be around 150–200BPM. If someone says he is dopesick but his pulse is under 100BPM then you can conclude that he was not being completely honest. If his pulse is sufficiently lower than normal then he probably just got high.

Testimony of acquaintances may be helpful. In my case, family members gave statements claiming that the deceased person seemed to be high in the days prior to his death and had been actively using for at least two weeks prior. They also stated that he seemed agitated the day prior but at one point went to an AA/NA meeting and came back in better spirits (factoid: opioid users who are actively using typically do not go to AA/NA meetings seeking fellowship). Agitation could suggest the onset of withdrawal, but it could also signify anticipation of withdrawal and frustration at being almost out of drugs (with a heavy user withdrawal will be obvious because it will involve vomiting, sweating, extreme restlessness, possibly screaming and/or crying, but with a lighter user it may be more subtle). The task of the defense, though, is not to establish anything definitively, which will likely not be possible, but simply to rebut the government’s confident assertions that the deceased person definitely was going through withdrawal and thus could not possibly have acquired drugs from anyone other than the defendant. It could be that he was buying from another source, for long enough to build up a tolerance, his regular source wasn’t responding and he started getting anxious about running out, he tried to use what he had left sparingly and took a Xanax bar or two (which might have had fentanyl in it) to lessen his anxiety, and then after he picked up from the defendant he was so relieved that he did all of what he had left from before in one shot and died. (Of course, in this scenario the defendant’s distribution would still be a but-for cause of death, but a conviction of distribution resulting in death requires the death to have resulted from use of the distributed drugs. See United States v. Hatfield, 591 F. 3d 945 (7th cir. 2010), for a detailed discussion of causation with respect to drug distribution cases.)

Putting It All Together

The stakes for a DRD trial are high, and even a partial win might not really be a win if the judge decides to use the acquitted conduct at sentencing (see United States v. Carvajal, 85 F. 4th 602 (1st cir. 2023)). A defendant with prior felonies will be looking at mandatory life if convicted. The jury will likely be tainted with misconceptions perpetuated by a sensational fear-mongering media. Oftentimes the defendant himself will know very little about what actually happened (in contrast to an actual murder case). Pleading guilty in most districts will result in a significant prison sentence, but losing at trial will, of course, result in a much longer sentence, possibly life. Some investigation will have to be done to supplement the government’s evidence and the defendant’s own recollections, after which a difficult decision will have to be made. Carefully analyze the deceased person’s phone records, the physical evidence, and the toxicology, and then at least the decision made will be an informed one. A long sentence is harder to cope with when you feel like it would have been shorter had you been more diligent.

Pleading Guilty

Here in the Central District of California the standard plea agreement is 12–16 years non-binding, non-negotiable, take it or leave it. The defendant must agree that level 38 is the correct base level, supported by both U.S.S.G. § 2D1.1(a)(2) and § 5K2.1 (which is untrue, since 5K2.1 explicitly requires sentencing judges to take into consideration the circumstances surrounding the death and the different levels of homicide). The government may or may not argue for a mass-marketing (§ 2D1.1(b)(7)) or misrepresentation (§ 2D1.1(b)(13)) departure. My lawyer argued for variances but no departures (which I think was a mistake) and the judge, Michael Fitzgerald, went a year under the guidelines (156 months) stating that he would have gone higher had there been any indication that the “victim” did not know he was buying fentanyl. Maybe the judge will take mitigating factors into consideration or maybe he won’t. But if an acquittal at trial seems unlikely, or if 20-life seems unpalatably risky, then all you can do is compile your mitigating factors, throw them at the judge, and hope for the best.

One possible departure that could be argued is § 5K2.13 Diminished Capacity. There is a disqualifier that this state of mind cannot have been due to the voluntary use of intoxicants, but withdrawal is a consequence not of use but of disuse. Moreover, repeated attempts by users to quit and failure to do so despite severe negative consequences (Alcoholics Anonymous calls this powerlessness; see Step 1 in Alcoholics Anonymous) calls into question the voluntariness of repeated use, even if initial use may have been voluntary. And repeated ingestion of opioids does severely diminish a person’s ability to control his inhibitions (and thus his ability to exercise sufficient willpower to cease using opioids). During undergrad I had no trouble forcing myself to go to sleep on time, get up early, and otherwise act like a responsible adult, but in the throes of opioid addiction it was an immense struggle just to force myself to do basic things like taking a shower everyday and showing up to classes on time.

Arguably, the sentence prescriptions for distribution resulting in death go against almost every subsection of 28 U.S.C. 994 (e.g. 994(m), to “develop a sentencing range that is consistent with the purposes of sentencing described in [18 U.S.C. § 3553(a)(2)]”). Most likely the plea agreement will stipulate that the defendant will not dispute that 38 is the correct base level, but that should not preclude arguing that, while it may be correct per the guidelines, the Sentencing Commission may have set the prescribed level higher than it ought to be (see Kimbrough v. United States, 552 U.S. 85, 128 S. Ct. 558, 169 L. Ed. 2d 481 (2007)). In effect, you will be agreeing to base level 38 and then arguing for a downward variance to reflect the inappropriateness of this prescription. You might reference the Sentencing Reform Act’s objective of “maintaining sufficient flexibility to permit individualized sentences when warranted” (28 U.S.C. § 991(b)(1)(B)), or the offense levels and elements associated with the various homicide offenses, or the huge disparities in sentences across districts as well as between state and federal offenses.

The core of my sentencing memorandum (disclaimer: the judge pretty much ignored all of our arguments and went with the guidelines before mumbling something about deterrence) was relative appreciation of risk. Basically, we were arguing from fundamental principles of culpability. The deceased person in my case was older than me, he had been using opioids for longer than I had been, he had convictions for drug charges, he was intentionally seeking “china”, and he chose to ingest his drugs by injection which I was not aware of (I always smoked my drugs). In contrast to the case routinely portrayed in the media of a high schooler taking a pill that he didn’t know had fentanyl in it there was no disparity of knowledge or experience, no pushy salesmanship, no “trickery”. The fact that someone died does not automatically render an offense “serious” (contrary to what the government will probably try to argue), otherwise involuntary manslaughter would not be level 12. We also cited studies claiming that “increases in severity of punishment do not yield significant (if any) marginal deterrent effects” (Michael Torry, Purposes and Functions of Sentences (2006)), and pointed out that unduly harsh sentences may promote fear of the law but not respect for it. We also mentioned the Sentencing Commission’s study on recidivism; although the zero-point offender departure is not available when death resulted from the offenses, the recidivism statistics for zero-point offenders are still relevant for arguing 3553(a) factors and downward variances.

In most situations, 18 U.S.C. § 3553(a) will favor the defense, because the draconian sentence prescriptions for distribution resulting in death go against all accepted principles of culpability and punishment (and the stated objectives of the Sentencing Reform Act). What disfavors the defendant is the guidelines, the propensity of judges to impose guideline sentences, and the stipulations that the defendant will be coerced into agreeing to due to the 20 year mandatory minimum of the initially charged offense. The government may agree to an 11(c)(1)© binding deal, but it probably will not be to the defendant’s liking. Sentencing statistics can also be procured from the Sentencing Commission on request (which encompass the entire range of possible prison sentences, from less than a year to life, and thus are a textbook example of “unwarranted disparities”).

RDAP and First Step Act

Defendants convicted of distribution resulting in death qualify for RDAP (Residential Drug Abuse Program) but not FSA. Defendants who plead guilty to simple distribution with a stipulation to having caused death in the factual basis should qualify for FSA but some institutions may say otherwise, in which case the administrative remedy process must be invoked. The most detailed and authoritative case right now is Lallave v. Martinez, 635 F. Supp. 3d 173 (E.D.N.Y. 2022). It is also worth noting that in 2023 the Sentencing Commission listed reconsideration of DRD sentencing as an item on its agenda (probably because of all the unwarranted disparities).

Motion to Acquit

I suspect (or hope, at least) that this charge may eventually make its way to the Supreme Court. The distribution resulting in death statute predates Apprendi, Booker, and Alleyne and blatantly contradicts a number of fundamental legal principles (see The Legal Dubiousness of Federal Overdose Charges). Consequently, a defendant intending to go to trial (or even someone planning to plead guilty) should consider putting together and filing a long-shot motion to acquit, or to dismiss the indictment. Such motions often focus on the lack of men’s REA, and one possible argument is that the imposition of strict liability for unintended conduct violates the Due Process clause of the Fifth Amendment. Another possible argument is that the mandatory sentence prescriptions, specifically mandatory life for a defendant with prior drug felonies, violates the Eighth Amendment prohibition against cruel and unusual punishment. My personal preference is for arguing that federally prosecution of homicide which did not take place in federal territorial jurisdiction violates the Tenth Amendment and the decision of United States v. Lopez, 514 U.S. 549, 115 S. Ct. 1624, 131 L. Ed. 2d 626 (1995), stating that the Commerce Clause does have limits and general police powers are state, not federal, powers. I like this approach because it asks a question that hasn’t been asked yet (Is this a distribution charge or is it a homicide charge?) whereas mens rea arguments have been tried and denied many times (e.g. United States v. Houston, 406 F. 3d 1121 (9th cir. 2005); United States v. Soler, 275 F. 3d 146 (1st cir. 2002)) (and also because I believe in federalism and would prefer to see this statute overturned on the grounds that state court is the proper venue for prosecuting homicides, even homicides that happen to have involved drugs).

Conclusion

Overdose charges in the feds are no joke. If you, or a client, or a loved one is considering going to trial then it is extremely important that you/he/she understand fully the situation. My aim was to provide some insights from my own experiences and research, but ultimately anyone facing a DRD charge will have to rely on his understanding of the particulars of his own case, together with the advice of his attorney. Wishful thinking, though, can be fatal, and if you choose to proceed to trial the verdict will be decided by a jury according to their understanding of the facts of the case (and their personal biases). But don’t let the prospect of trial scare you into forgetting that pleading guilty also carries uncertainty and risk. Sometimes in life a good outcome is not possible and all that you can do is try to keep the damage to a minimum. Hopefully I was able to provide a little bit of thought provoking insight to anyone in this situation. Best of luck.