The Best Treatment for Stimulant Addiction Pays You to Stay Clean. For Forty Years, We Refused to Try It.

Somebody checks a urine sample, it’s clean, and they hand over a voucher worth somewhere between one and a hundred dollars — a gift card, a shot at a prize drawing, sometimes just cash. That’s it. That’s the entire mechanism of contingency management, the most effective treatment we have for stimulant use disorder, and the reason most people have never heard of it is that for four decades, the country’s answer to stimulant addiction wasn’t treatment at all. It was a prison sentence, and the length of that sentence depended on which form of the drug you were caught holding.

We had a treatment that works for stimulant addiction since the 1990s. We spent the years in between building prisons instead of paying for it.

What contingency management actually is, and why it works

Strip away the clinical name and contingency management (CM) is exactly what it sounds like: you get rewarded for evidence that you didn’t use. A negative drug test, an attended appointment, a documented step toward a goal you set — each one triggers a small, immediate incentive. Research summarized by the University of Pennsylvania’s Leonard Davis Institute finds CM roughly twice as effective as cognitive-behavioral therapy, counseling, or motivational interviewing alone for stimulant use disorder, with effects that can persist at least a year after treatment ends when incentives are structured well — researchers recommend $100 to $200 a month for optimal effect. The Department of Veterans Affairs has run CM programs in more than 100 medical centers since 2011, treating over 6,300 veterans, with more than 90% of participants testing negative for their target substance during the program.

The reason it works isn’t mysterious, and it’s worth saying in plain terms: unlike opioid use disorder, there is no FDA-approved medication that directly treats methamphetamine or cocaine addiction the way buprenorphine treats opioid addiction. Stimulants act on dopamine and reward circuitry in ways that behavioral, not pharmacological, interventions currently address best — and CM works by directly engaging that same reward system, offering a small, certain, immediate reward to compete against the drug’s larger but riskier one. It’s not a metaphor for treating addiction. It’s using the addiction’s own mechanism, redirected.

The gap between the evidence and the funding

None of this is new science. Early contingency management trials for stimulant use disorder date to the 1990s. What’s new is that Medicaid — the payer that actually reaches the population most affected by stimulant use disorder — is only now starting to cover it. California’s Medi-Cal Recovery Incentives program, approved in 2021, caps benefits around $599 a year, well below the $100–200 monthly level researchers recommend for full effect, but it’s real coverage where none existed before. Washington followed in 2023. Delaware, Montana, and West Virginia have requests pending. That’s five states, thirty-some years after the first trials, for a treatment with a better evidence base than most things Medicaid already pays for without a second thought.

A new cohort study in the American Journal of Psychiatry gives the funding gap sharper stakes: it links contingency management for stimulant use disorder to a measurable reduction in mortality among participants, not just reduced use — meaning the years states spent not covering CM weren’t a neutral policy choice. They were years with a preventable death toll attached, the same way the years before naloxone went over the counter were. The study’s authors are appropriately careful about certainty here — a cohort design shows association, not proof of causation the way a randomized trial would, and the population studied skews toward people already engaged enough with a treatment system to be enrolled in the first place. That caveat doesn’t erase the finding. It just means the honest version of the claim is “this is linked to fewer deaths in the population we could study,” not “this saves lives, full stop” — and a clinician should trust a study more, not less, for saying so plainly.

They were years with a preventable death toll attached, the same way the years before naloxone went over the counter were.

Why stimulant users got prison instead of treatment

To understand why it took thirty years to fund a treatment that worked from the start, it helps to look at what the country built instead. The Anti-Drug Abuse Act of 1986 set a 100-to-1 sentencing disparity between crack and powder cocaine — five grams of crack triggered the same five-year mandatory minimum as 500 grams of powder, despite the two being nearly chemically identical. Congress built that disparity on the theory that crack was uniquely violent and uniquely addictive; the U.S. Sentencing Commission later concluded the premise was wrong and the racial impact was severe. It took until 2010 for the Fair Sentencing Act to narrow the ratio to 18-to-1 — still not equal, just less unequal — and the sentences already served were not undone for most of the people who served them. That is the infrastructure this country built for stimulant use disorder while contingency management sat proven and unfunded: not a treatment system, a sentencing schedule.

The new risk CM has to compete against

Contingency management also matters for a reason that has nothing to do with 1986 and everything to do with this month: the stimulant supply itself has changed. Community drug-checking data shows fentanyl now turns up in roughly 6% of methamphetamine samples and nearly 13% of cocaine samples — contamination that has nothing to do with the person’s intent to use an opioid and everything to do with a supply chain that doesn’t separate its product lines. That’s the harm-reduction case for CM sitting alongside the treatment case: every negative drug screen a CM program rewards is also, functionally, one fewer roll of the dice against a supply that can kill someone who never meant to touch fentanyl at all.

If you use stimulants, or you’re the one testing negative for a voucher this week, the incentive on the table is real money for a real result — not a trick, not a gimmick, a structure built on the same reward pathway the drug already uses, aimed the other direction. The country is only now getting around to paying for it. That’s not a reason to wait for your state to catch up before it’s worth doing. It’s a reason to ask, directly, whether your local program already qualifies for the Medicaid codes that exist right now — because five states already say yes, and the sixth one might be closer than it looks from here.