The Autopsy / Case File · Overdose & Substance
The Other Charge
As of July 2025, every state in the country, plus the District of Columbia and the Northern Mariana Islands, has a Good Samaritan law promising that a person who calls 911 during an overdose won't be arrested for it. Devan Miller called 911. She was arrested anyway — for a charge the law was never written to cover.
The Legislative Analysis and Public Policy Association's 2025 fifty-jurisdiction tracking confirms the headline achievement of a decade of drug-policy advocacy: every U.S. state now grants some form of limited criminal immunity to bystanders who summon emergency help for an overdose. What that tracking also documents, jurisdiction by jurisdiction, is the boundary of that immunity — a boundary that stops well short of the charge prosecutors actually reach for in a growing number of fatal-overdose cases. Drug-induced homicide statutes, now on the books in more than 30 states plus DC, Guam, and the U.S. Virgin Islands, sit entirely outside Good Samaritan protection. The Health in Justice Action Lab's review of media-reported cases found 3,264 such prosecutions between 1974 and 2022, with the sharpest rise tracking fentanyl's spread — and found that most of those charged were low-level dealers or, more often, someone with a personal relationship to the person who died: a friend, a partner, the person who was there.
Join the discussion →
The design logic behind a Good Samaritan overdose law is straightforward and, on its own terms, sound: the single biggest documented barrier to a bystander calling 911 during an overdose is fear of arrest, so remove the arrest risk for the specific act of calling, and more people call. Every state has now tested that logic at scale. LAPPA's July 2025 summary of state laws — the most current comprehensive tracking of the policy — confirms that all 50 states, the District of Columbia, and the Northern Mariana Islands have adopted some version of it, granting limited protection from criminal liability to both the person who calls for help and, in most jurisdictions, the person who overdosed. It's the closest thing American overdose policy has to a settled, near-universal consensus. And it comes with a boundary line that the underlying research on 911-calling behavior shows most bystanders don't actually understand until the moment it matters.
What "Immunity" Actually Covers
Good Samaritan overdose statutes were written to solve one specific problem: low-level drug possession and paraphernalia charges that would otherwise attach to anyone present at an overdose scene, caller and victim alike. That's what the immunity covers, and states vary widely in how far even that narrower protection extends. Wyoming caps its protection at two overdose calls in a 12-month period, and only grants the second one if the person agrees to enter an approved treatment program. Iowa and South Dakota grant it exactly once. Ohio allows two uses, contingent on the person completing a substance-use screening and treatment referral within 30 days. Texas denies the defense entirely to anyone who has already used it once, or who has called emergency services for an overdose within the preceding 18 months — a provision that, read plainly, means the second overdose a person witnesses in a year and a half is the one where calling 911 carries real legal risk again. And in the large majority of jurisdictions, the immunity explicitly does not limit a probation or parole officer's authority to drug-test a person on supervised release, meaning a call that avoids one criminal charge can still trigger a probation violation.
None of that is the gap this piece is about. Those are all still possession-and-paraphernalia-adjacent limitations, operating inside the category of charge Good Samaritan laws were designed to touch. The gap that matters more, and that the laws were never designed to touch at all, sits one level up the charging ladder.
The Charge the Law Was Never Written to Cover
Drug-induced homicide (DIH) laws, sometimes styled as "drug delivery resulting in death" statutes, let a prosecutor charge the person who supplied, shared, or was otherwise present when a lethal dose changed hands with something on the order of manslaughter or murder — penalties that in some states run as high as 60 years. The Health in Justice Action Lab, which has built the most comprehensive independent tracking of these cases from court records and media reports, counts more than 30 states plus DC, Guam, and the U.S. Virgin Islands with an express DIH or drug-delivery-resulting-in-death statute on the books, and finds that all 50 states have some legal mechanism — DIH statutes, felony murder rules, or general homicide statutes stretched to fit — that a prosecutor can use to charge someone in connection with a fatal overdose. Good Samaritan immunity, in every one of these jurisdictions, is silent on all of it. The statutes were drafted to remove the possession charge, not the homicide charge, and no state's Good Samaritan law purports to grant a caller protection from a prosecutor who decides the caller was also the source of the drug.
That distinction is not hypothetical, and it is not rare enough to dismiss as an edge case. In Illinois, Devan Miller called 911 after her friend overdosed. She was arrested and charged with drug delivery — a charge Illinois's own Good Samaritan statute does not cover, because prosecutors alleged she had helped administer the drug, which under Illinois law moves the case from "possession, protected" to "delivery resulting in death, not protected" in a single allegation. Her case became a widely cited example inside harm-reduction and public-defense circles specifically because it demonstrates, concretely, what the immunity's boundary looks like from the inside: you can be fully protected from the possession charge and still be fully exposed to the homicide charge, in the same case, over the same overdose, because you called for help.
You can be fully protected from the possession charge and fully exposed to the homicide charge — in the same case, over the same overdose, because you called for help.
Who the 3,264 Cases Are Actually About
The scale and shape of DIH prosecutions matter as much as their existence. The Health in Justice Action Lab's dataset of 3,264 identified cases spanning 1974 through 2022 shows a prosecution rate that tracked closely with the rise of fentanyl in the drug supply, concentrated disproportionately in the Midwest and Northeast. But the more consequential finding is about who actually gets charged. The research describes a majority of prosecutions as targeting low-level dealers or people with a personal connection to the deceased — not the traffickers or manufacturers the "homicide" framing implies, but often the boyfriend, the roommate, the using partner who was in the room, sometimes the same person who dialed 911. The dataset also documents a racial disparity in who bears that prosecutorial weight, consistent with broader patterns in the criminal legal system's handling of drug offenses. None of that population is who the Good Samaritan statutes' drafters were picturing when they built an immunity around "the caller."
Why the Boundary Undoes the Law's Own Purpose
The research on why bystanders hesitate to call 911 in the first place makes the DIH gap more than a technical inconsistency — it makes it a direct undermining of what Good Samaritan laws exist to accomplish. A study following people trained in overdose response found that only 2 of 13 documented interventions resulted in a 911 call placed by the responder; two more calls were made by someone else at the scene, not the trained bystander. The dominant reason people gave for not calling wasn't fear of a possession charge at all — it was fear that a police response would trigger an identification check leading to an outstanding-warrant arrest, or a parole or probation violation, for either the caller or the person overdosing. None of the housed participants with subsidized housing called 911 in the overdoses they witnessed, a pattern researchers tied directly to federal rules that allow Section 8 housing to be terminated over drug activity on the property even without a conviction — a consequence a Good Samaritan possession-charge immunity does nothing to prevent, because losing housing isn't a criminal charge. Participants also described a specific, evidence-based skepticism about whether the legal promise would hold up in practice: "a legacy of mistrust" between what the law says on paper and what an individual officer does on scene.
Layer the DIH gap onto that picture and the incentive structure gets worse, not better, for exactly the people most likely to be present at an overdose: the people who use drugs together, who sometimes share or split a supply, and who are the most likely bystanders precisely because they're in the room. The Health in Justice Action Lab's research states the mechanism plainly — DIH prosecutions discourage witnesses from calling 911 for fear of being charged with DIH or another serious crime, directly eroding the effectiveness of the Good Samaritan protections that were supposed to solve that exact fear. A law built to remove one specific deterrent to calling for help coexists, in the same 30-plus states, with a charge built to reintroduce a bigger version of that same deterrent — aimed at the people the caller is most likely to be.
A Protection With a Ceiling Nobody Advertises
None of this means Good Samaritan laws don't work as far as they go — the research on 911-calling behavior does show that correct knowledge of a state's law is associated with a higher likelihood of calling, when people actually understand what it covers. The problem is the gap between what the public message says ("call, you're protected") and what the statute itself actually promises, a gap that a caller like Devan Miller discovers only after the police arrive and start asking who provided the drug. Forty-eight or fifty states advertising overdose immunity is a genuine, hard-won policy achievement. It is also, in every one of the 30-plus states with a live drug-induced homicide statute, a promise with a ceiling that isn't printed anywhere near the headline.
And Then You're Dead
The overdose death a Good Samaritan law is built to prevent is often not the one that ends the story. Devan Miller's friend died. Devan Miller called for help, and got arrested for the same event — a sequence the law that was supposed to protect her was never actually built to stop, in the more than 30 states where that sequence remains entirely legal.
That's it. That's the whole thing.
Sources
- All 50 states, DC, and the Northern Mariana Islands having a Good Samaritan fatal-overdose-prevention law as of July 2025, and jurisdiction-by-jurisdiction limits (Wyoming's two-use/12-month cap, Iowa and South Dakota's one-time limit, Ohio's two-use-plus-referral condition, Texas's 18-month exclusion, and the general non-limitation of probation/parole drug-testing authority) — Legislative Analysis and Public Policy Association (LAPPA), "Good Samaritan Fatal Overdose Prevention and Drug-Induced Homicide: Summary of State Laws," July 2025
- More than 30 states plus DC, Guam, and the U.S. Virgin Islands having an express drug-induced homicide (DIH) or drug-delivery-resulting-in-death statute; the 3,264 identified DIH prosecutions from media reports, 1974–2022; the concentration of prosecutions in the Midwest and Northeast; the finding that most prosecutions target low-level dealers or people with a personal connection to the deceased rather than traffickers; the documented racial disparity in prosecution; and the finding that DIH prosecutions discourage 911 calling and undermine Good Samaritan law effectiveness — Health in Justice Action Lab, Northeastern University School of Law, "Drug Induced Homicide as a Response to Overdose"
- The Illinois case of Devan Miller, charged with drug delivery after calling 911 to help an overdosing friend, as an example of Good Samaritan law's exclusion of anyone alleged to have supplied or administered the drug — reporting cited by Illinois harm-reduction and criminal-defense legal analysis of Illinois's Good Samaritan statute
- The study of trained overdose responders finding only 2 of 13 documented interventions resulted in a 911 call by the trained bystander, the role of fear of an identification check leading to arrest for outstanding warrants or parole/probation violations, the finding that no housed participants called 911 due to Section 8 housing termination risk from drug activity even without conviction, and the documented distrust that a Good Samaritan law's protection would be honored in practice — peer-reviewed qualitative research on overdose-witness 911-calling behavior, published in the International Journal of Drug Policy
- Photo: "Handcuffs on table," U.S. Air Force photo by Airman 1st Class Gustavo Castillo, public domain, via Wikimedia Commons
Comments
Loading…