There's Still No Pill for Meth Addiction. There's a Gift Card, and It's Working.

Bernard Groves went to lunch with his aunt in the summer of 2024 and watched her face while they talked. “I went [to lunch] with my auntie and I saw such sadness in her eyes,” he told NPR. He’d spent five years by then cycling in and out of meth relapse. What got him to that lunch clean wasn’t a pill — there isn’t one — it was a program that gave him a Visa gift card every time he peed in a cup and tested negative.

The single most effective medical tool against methamphetamine and cocaine addiction in America right now is not a medication. It’s a reward.

That’s not spin. It’s the state of the science. Stimulants — methamphetamine and cocaine combined — were involved in 59% of all U.S. overdose deaths between January 2021 and June 2024, according to the CDC’s own count: 182,502 deaths, a rate that has climbed for a decade. And unlike opioid use disorder, which has three FDA-approved medications (methadone, buprenorphine, naltrexone), stimulant use disorder has zero. The FDA didn’t even issue its first formal guidance for pharmaceutical companies on how to design a stimulant use disorder drug trial until October 2023 — an acknowledgment, decades into the crisis, that the industry hadn’t been building toward an answer because no one had told it what an answer needed to look like.

Into that vacuum stepped contingency management — psychologist-speak for “pay people when they don’t use.” It sounds too simple to be medicine, and Groves said as much himself, laughing at his own turnaround: “Like, how could you say you’re excited to pee in a cup? But I was.”

What craving actually is, and why a gift card can out-argue it

If you’ve used meth or watched someone you love use it, you already know that “just stop” was never realistic advice — the pull back to the drug isn’t a character flaw, it’s a measurable brain state. A 2026 study in Frontiers in Public Health tracked 134 people in a compulsory rehabilitation setting and found something specific: the craving that best predicted how strongly someone’s attention got hijacked by drug-related images wasn’t the sharp urge triggered by seeing a pipe or a picture. It was a quieter, more constant “withdrawal craving” — the background pull that persists for weeks after the drug is gone, independent of any specific trigger. Researchers call the phenomenon “attentional bias”: your brain starts treating drug cues the way it treats a fire alarm, impossible to fully tune out, even when you’re consciously trying to look away.

Contingency management works, researchers believe, because it gives that background craving something to compete against in the same currency the brain already understands: immediate, tangible reward. Rick Rawson, professor emeritus of psychology at UCLA and one of the method’s earliest champions, remembers how badly the idea was received when he started making the case decades ago: “You would hear things like, this isn’t treatment, this is just paying people,” he told NPR. For years, he says, “I’d pretty much given up… I figured this just isn’t going to happen.”

It happened anyway, because the data kept winning arguments the politics couldn’t. California’s Medicaid contingency management program had grown to nearly 4,000 participants by September 2024, with at least 75% of urine samples testing negative — gift cards running $10 to $26.50 per clean test, capped around $599 over six months. Groves used his on things that sound almost aggressively ordinary: exercise weights from Walmart, bird food for his pet, a movie night funded for his mother, sister, and grandmother. “Being able to treat my family and do things for them is special,” he said. “It brought some joy back in my life.”

The gap the model doesn’t close is access. Ayesha Appa, an addiction specialist who runs an HIV clinic at San Francisco General Hospital, told NPR about a homeless patient on Medicaid who died of an overdose before contingency management became available outside specialty clinics near her. “It feels both incredibly frustrating and just heartbreaking as a provider,” Appa said. “What if we could have offered her contingency management in the clinic… already?” California, Montana, Washington, and Delaware currently run Medicaid-funded contingency management; most states, including Arizona, don’t yet. If you’re a case manager working stimulant-use caseloads in a state without a Medicaid CM benefit, the question worth raising at your next team meeting isn’t whether the evidence is strong enough — it is — it’s which of your current program’s incentive structures (attendance stipends, transportation vouchers, housing-first supports) could be restructured around the same behavioral principle without waiting on a state waiver.

“What if we could have offered her contingency management in the clinic…

Arizona’s stimulant problem has a schedule, and it’s July

Arizona doesn’t need to import a stimulant crisis to understand why this matters — it has one on its own calendar. Maricopa County’s fatal overdoses run more than double January’s rate every July, and it isn’t random: stimulants dehydrate users and blunt the body’s ability to regulate its own temperature, a dangerous combination during a Phoenix summer. Last year, substances were a factor in 55% of the county’s heat-related deaths, and stimulants were involved in roughly 8 of every 10 of those. Ariella Dale, chief science officer at the Maricopa County Department of Public Health, has been urging residents to keep naloxone on hand through the hottest months — a reminder that even a stimulant-driven overdose can involve fentanyl-contaminated supply, and naloxone can still matter.

The government has punished stimulants unevenly before

This isn’t the first time American drug policy has struggled to treat stimulants as a medical problem instead of a moral one, and the last time offers a warning about what happens when it doesn’t. In 1986, at the height of the crack cocaine panic, Congress set a 100-to-1 sentencing disparity between crack and powder cocaine — the same drug, chemically, sentenced as though it were a hundred times more dangerous depending on which form showed up in a defendant’s pocket. The Fair Sentencing Act of 2010 cut that ratio to 18-to-1, not to zero, and by then the damage to a generation of overwhelmingly Black defendants was done: 81% of people convicted of crack offenses in 2007 were Black, in a country where a quarter of crack users were. The lesson wasn’t that cocaine wasn’t dangerous. It was that the policy response was never actually calibrated to the pharmacology — it was calibrated to who policymakers pictured using it.

Today’s stimulant response looks different on its face — CDC data, FDA guidance documents, Medicaid pilot programs — but the same instinct that built the 100-to-1 ratio still shows up in subtler forms: in which states fund contingency management and which don’t, in whose stimulant use gets met with a gift card program and whose gets met with a cell. The CDC’s own 2026 mortality data shows stimulant death rates climbing fastest among Native American and Black populations, the same populations the 1986 law fell heaviest on.

There is no pill for what Bernard Groves was fighting, and there may not be one for years. What exists — a gift card, a peer, a program that treats staying alive as worth rewarding rather than punishing — is not a lesser substitute for medicine. Right now, for stimulants, it’s the best medicine there is.