And Then You're Dead

The Autopsy / Case File · Overdose & Substance

The Data Nobody Runs

A drug-checking service can tell someone what's in their bag in minutes. The federal government's own mortality count takes six months to catch up.

There is no national agency whose job is to know, in real time, what's actually in the American drug supply. What exists instead is a scattered network of syringe service programs, university labs, and harm-reduction nonprofits running mass spectrometers and test strips on their own budgets, feeding a surveillance system that federal agencies lean on but don't operate. Here's how the fastest overdose-prevention data in the country gets made, and why it keeps almost losing its funding.

And Then You're Dead · September 2026 · 8 min read

Join the discussion →
A handheld Raman spectrometer, the class of portable device used to identify unknown chemical samples in the field
A handheld Raman spectrometer. Devices like this — alongside infrared spectrometers and immunoassay test strips — are what drug-checking programs use to tell someone what's actually in their supply, often before the drug ever reaches a body.

When someone dies of a drug overdose in the United States, the death eventually gets coded, filed, and folded into the CDC's official count. That count is the number news outlets cite, the number that shapes federal funding formulas, the number the country uses to ask whether the crisis is getting better or worse. It is also, by the CDC's own admission, a number that lags reality by months — long enough that a new adulterant can spread through a regional drug supply, kill people, and still not show up in the government's own numbers until well after the fact.

A Six-Month Blind Spot, By Design

The National Vital Statistics System, which produces the country's official overdose death counts, runs on death certificates — documents that have to be completed by a physician or medical examiner, coded by a state vital records office, and transmitted to the National Center for Health Statistics before a death is countable at all. Toxicology results alone, in cases that require them, can take weeks. The result is a reporting lag of roughly six months between when a fatal overdose happens and when it's reflected in CDC's provisional data, with fully finalized annual counts arriving later still.

That lag isn't a flaw anyone is trying to hide. CDC publishes the provisional numbers with the caveat attached, and adjusts for the undercount using statistical modeling once enough time has passed for the true rate to become clear. But a statistical correction, however sound, doesn't help a person deciding whether it's safe to use a specific bag from a specific supply tonight. For that, the six-month system is functionally useless. Something else has to fill the gap.

0 mo.
approximate reporting lag between a fatal overdose and its appearance in CDC's provisional National Vital Statistics System count
0
federal agencies whose statutory mission is to test the street drug supply in real time and report results back to the public

What A Test Strip Can Tell You In Ninety Seconds

The gap is filled, unevenly, by drug-checking services — syringe service programs, university research partnerships, and standalone harm-reduction nonprofits that take a sample of a person's actual drugs, run it through a fentanyl test strip or a portable spectrometer, and hand back a result before that person uses. It is, structurally, the opposite of the federal system: instead of a national count built retroactively from death certificates, it's a hyperlocal, real-time readout of what a specific supply actually contains, built one voluntary sample at a time.

A study led by researchers at Rhode Island Hospital, funded through the Bloomberg American Health Initiative at Johns Hopkins Bloomberg School of Public Health, tested this approach directly: more than 200 samples of police-confiscated street drugs were run through three portable technologies — a benchtop Fourier-transform infrared spectrometer, a handheld Raman spectrometer, and fentanyl test strips — to see which actually worked in the field. The test strips, the cheapest and simplest of the three by a wide margin, came out on top for detecting fentanyl itself.

0%
sensitivity range for fentanyl test strips in detecting fentanyl and its analogs, across the two most accurate devices tested in the Rhode Island Hospital/Johns Hopkins field study
0 ng/mL
detection limit of laboratory-grade liquid chromatography–tandem mass spectrometry, the gold-standard method some drug-checking labs use to screen for 22 common drugs and metabolites at once

The tradeoffs are real. Fentanyl test strips are cheap and fast but tell a user only whether fentanyl is present, not how much, and not what else might be mixed in. Spectrometers can identify a much broader range of substances but cost thousands of dollars, require some training to interpret, and can struggle with mixtures where one compound masks another. Laboratory-grade mass spectrometry catches almost everything but requires shipping a sample off-site and waiting for results — a turnaround measured in days, not the minutes a person standing outside a syringe service van actually has.

No test strip and no spectrometer can move faster than a six-month government lag needs it to. The problem drug-checking programs actually solve is a different one: getting a real answer to a real person before they use, not building a better national count after the fact.

A Legal Patchwork Built On Top Of A Data Patchwork

Even where the technology works, the law hasn't consistently caught up to it. For years, most states classified fentanyl test strips as drug paraphernalia — the same legal category as the syringes and pipes explicitly meant to be illegal to distribute — which meant a harm-reduction worker handing one out could, in some jurisdictions, be committing a crime. That changed only gradually: in 2021, CDC and the Substance Abuse and Mental Health Services Administration announced that federal harm-reduction grant money could finally be used to purchase fentanyl test strips, a policy shift that acknowledged what public health researchers had been arguing for years.

0
states that prohibited possession of fentanyl test strips as drug paraphernalia before the 2021 federal policy shift; 33 prohibited their distribution
0
categories of test strip most states have now exempted from paraphernalia law — fentanyl strips specifically. Strips for other adulterants, including benzodiazepines and xylazine, generally remain uncovered

Ohio's 2022 update to its paraphernalia statute is a representative example of how narrow these fixes tend to be: Senate Bill 288 carved out an exception specifically for fentanyl test strips, and nothing else. A harm-reduction program that wants to hand out test strips for xylazine, benzodiazepines, or any of the other adulterants now circulating alongside fentanyl is often working in the same legal gray zone fentanyl strips occupied a few years ago — waiting on a state legislature to notice a specific product exists and write it a specific exemption, one substance at a time, while the drug supply itself keeps moving faster than that process does.

Who's Actually Running The System That Works

No single organization coordinates drug checking nationally. The closest thing to a coordinating body is the National Drug Early Warning System, a sentinel-site network originally built through a cooperative agreement between the National Institute on Drug Abuse and the University of Maryland's Center for Substance Abuse Research. NDEWS doesn't run drug-checking devices itself; it aggregates trend data from sentinel sites and community-based experts around the country and treats venue-based drug-checking results as one input among several, feeding a public health early-warning function that is explicitly meant to move faster than official mortality statistics.

The actual testing work happens downstream of that, inside individual syringe service programs and university partnerships that vary enormously by state and by budget. CDC's Overdose Data to Action initiative requires a subset of its funded state and local health department grantees to run drug-checking or paraphernalia-testing programs as one surveillance strategy among many, which means federal money does flow into this work — but through a grant condition attached to a broader initiative, not through an agency whose sole job is running a national drug-checking network. A 2024 implementation study of programs in Ohio and Indiana, published in Harm Reduction Journal, found individual syringe service programs cobbling together funding from university research grants, state opioid settlement dollars, and philanthropic sources like the Berthiaume Institute for Precision Health — a funding base that can vanish with a single grant cycle in a way NVSS's federal budget line never does.

0
low-cost, low-barrier drug-checking methods evaluated for syringe service program use in the 2024 Ohio/Indiana implementation study — colorimetric paper tests, absorption tests, reagent-tube tests, and repurposed rapid immunoassay strips
0
year NIDA first funded the National Drug Early Warning System's sentinel-site network at the University of Maryland, explicitly to catch emerging drug threats faster than mortality data could

That funding precarity is the structural problem underneath the technical one. Drug-checking services have already solved the speed problem the federal mortality-counting system can't solve by design — a test strip doesn't need six months to tell someone what's in their bag. What they haven't solved, and can't solve on their own, is durability. A syringe service program running on a two-year settlement grant can build exactly the real-time surveillance capacity the country's official numbers lack, and then lose it the moment that grant isn't renewed — while the six-month lag in the federal count, underfunded as its own surveillance infrastructure sometimes is, keeps running every year on autopilot because it's written into a permanent statistical system. The fastest data in the overdose crisis is also the most fragile.

And Then You're Dead

A test strip works in ninety seconds. A federal mortality count takes six months. The country built the slow system to last and the fast one to depend on whichever grant cycle happens to still be funded this year — and the gap between those two timelines is measured in overdoses nobody saw coming until the paperwork caught up.

That's it. That's the whole thing.

Comments

Loading…

Comments are anonymous and moderated.

Sources

Photo: "Portable RAMAN Laser Spectrometer" by Dean Calma/IAEA Imagebank, via Wikimedia Commons (CC BY 2.0), cropped from the original.