A Gift Card Cut Overdose Deaths by 41%. The VA Data Just Proved It Wasn’t a Fluke.

For years, the strongest evidence behind contingency management — the practice of paying people small, escalating rewards for biochemically verified abstinence — came from controlled trials, the kind run on a few hundred motivated volunteers under close supervision. Critics could always say the same thing: nice results, but that’s not what happens in a real clinic, with real caseloads, with people who didn’t sign up for a study.

A national cohort study using linked Veterans Health Administration records, tracking patients from mid-2018 through the end of 2020, just took that objection off the table. Veterans with stimulant use disorder — methamphetamine and cocaine, the two substance classes with zero FDA-approved medications between them — who received contingency management in routine VA care were 41% less likely to die within a year than matched patients who didn’t. This wasn’t a lab. This was ordinary care, at scale, with the messiest population medicine treats.

Meth doesn’t have a medication. It has never needed one to save this many lives — it needed a system willing to pay for the thing that already works.

If you’ve been told methamphetamine use disorder is essentially untreatable because there’s no pill for it, that framing has always been true and always been misleading in the same breath. There’s no medication because stimulant pharmacology hasn’t produced one that clears the FDA’s bar — a real, unresolved scientific gap. But contingency management was never waiting on that gap to close. It rewards the behavior directly: a clean test, a small payment, repeated and escalated, building toward the kind of stability medication provides for opioid use disorder by a completely different route. The NEJM Clinician summary of the real-world evidence calls this the clearest mortality signal contingency management research has produced to date — not reduced use, which is the outcome most studies chase, but reduced deaths, which is the only outcome that actually matters to the person living it.

The gap now isn’t scientific. It’s regulatory and financial: federal anti-kickback statutes were written before “structured incentive payments” was a recognized clinical intervention, and untangling that took years most state Medicaid programs haven’t caught up on yet. Only a handful of states currently cover contingency management under Medicaid. Every veteran in this study who got it did so inside a federal system that had already solved the coverage question for its own population — which means the barrier other states are still working around isn’t whether contingency management works. The VA already answered that. It’s whether the rest of the country is willing to act like it heard the answer.