Arizona’s Meth Deaths Spike Every July. The State Still Hasn’t Applied for the Treatment That Works.

It’s July in Maricopa County, and the county’s chief science officer, Ariella Dale, is explaining to KJZZ why this month kills more people who use drugs than any other. Methamphetamine, she says, “makes us more dehydrated, it’s harder to regulate your temperature, it increases your heart rate.” Every one of those effects gets worse once the air itself is 115 degrees. July has been the county’s deadliest overdose month for years running — roughly double January’s rate, the worst month on record landing during the hottest July on record. Stimulants were involved in 8 of the county’s last 10 heat-related overdose deaths. This isn’t a mystery playing out once. It’s a mechanism, and it runs on the calendar.

Arizona hasn’t applied for the one Medicaid waiver that funds the treatment proven to work on the drug killing its residents every summer.

That treatment is called contingency management — plain language, it means paying people small, structured rewards, like gift cards or vouchers, for verified drug-free test results, the same way a workplace wellness program might reward you for hitting a step count. It sounds almost too simple to be real medicine. It is real medicine. It’s also the only thing standing between Arizona’s July body count and a treatment system built to actually meet it.

No pill treats this. One behavioral treatment does, and the newest research just made the case for it stronger

Substance use factored into 55 percent of the county’s heat-related deaths last year, and stimulants drove nearly all of them. Start with the plain fact that makes this whole story necessary: in 2026, there is still no FDA-approved medication for methamphetamine use disorder or cocaine use disorder. None. Opioid use disorder has buprenorphine and methadone. Alcohol use disorder has naltrexone. Stimulant use disorder has nothing pharmaceutical to offer. A JAMA Psychiatry meta-analysis of 12 federally funded randomized controlled trials, published June 3, 2026, confirmed it again: across roughly 2,000 participants, “no medication demonstrated a significant benefit over placebo” when the yardstick was total abstinence.

But the same researchers found something more useful buried in that null result. When they re-scored the same trials using a “reduced use” outcome — not “did this person stop entirely,” but “did this person use meaningfully less” — the effect sizes got consistently larger, and one drug, cabergoline, showed a real signal for reduced cocaine use that abstinence-only scoring had erased. That’s the layman version of what researchers call an abstinence-only endpoint: a way of grading treatment that only counts a total win, and throws out everything short of it, even when “using four days a week instead of seven” is the difference between a person keeping their job and losing it. The study’s authors argue for grading stimulant treatment on a continuum instead. That argument matters for contingency management too, because CM’s own decades of evidence — more than 200 randomized controlled trials at this point — has always been strongest on exactly that continuum: fewer positive screens over time, more days engaged in treatment, more weeks without a relapse severe enough to land someone in the ER.

Contingency management isn’t new or untested. It grew out of decades of behavioral psychology research into operant conditioning — the study of how consequences shape behavior — and psychologist Stephen Higgins adapted it for cocaine use disorder in the early 1990s at the University of Vermont, building voucher-based incentive programs that outperformed standard counseling trial after trial. For thirty years afterward, it stayed a conference favorite and a real-world afterthought, because the objection was never really about the data. It was moral. Regulators capped incentive payments at $75 a year, and as one addiction-counselor training resource summarizing that era put it, officials were “frightened of approving any sort of program” that looked like handing public money to people struggling with drugs. That cap held for roughly two decades before it moved.

For thirty years afterward, it stayed a conference favorite and a real-world afterthought, because the objection was never really about the data.

If you’re the one this is actually about — using meth or cocaine, or trying to stop, and tired of being told to just want it more — the science says your cravings and your low points aren’t a character problem. They’re the exact thing operant conditioning was built to work with. A negative screen earning you something real, right now, isn’t a gimmick; it’s giving a reward system that’s been hijacked by the drug a legitimate, achievable win to chase instead. CM meets your brain where it already lives instead of arguing with it.

Five states now pay for this. Arizona spent $582 million on treatment and never asked.

In January 2025, SAMHSA finally moved on the money question: a federal advisory raised the incentive cap from $75 a year to $750 a year, still non-cash, still restricted from being spent on alcohol, tobacco, cannabis, or lottery tickets, but ten times the buying power to work with. That single regulatory change is arguably the biggest thing to happen to stimulant treatment policy in a decade, because $75 a year buys almost nothing — a couple of gas cards, maybe a month of bus passes — while $750 can fund the kind of escalating-value voucher schedule that the original research actually tested. Programs no longer have to pretend a token gesture is a real incentive. They can build one.

Medicaid coverage is the other lever, and it moves state by state. Under Section 1115 waivers — special federal permission slips that let a state’s Medicaid program pay for services it wouldn’t otherwise cover — California, Delaware, Hawaii, Montana, and Washington now have approved CM waivers. Rhode Island’s went into effect January 1, 2026, projected to serve about 575 people in year one and roughly 1,150 a year after that. Michigan’s application is pending. KFF’s tracker estimates that if every pending waiver clears, roughly 26 percent of the Medicaid population nationally diagnosed with a stimulant use disorder would have coverage for the treatment that actually works for their diagnosis. Roughly one in four. That’s the ceiling right now, and it’s still a floor Arizona hasn’t reached, because Arizona has not applied for a waiver at all.

Meanwhile, AHCCCS’s own numbers make the absence harder to explain away. The agency’s FY2025 annual substance use treatment report shows methamphetamine and other stimulants were the third most common diagnosis among the 143,997 Arizonans who received SUD treatment last year — 45,117 people, trailing only alcohol and opiates — while the state spent $582,342,402 total on substance use treatment services. And the death data isn’t holding steady while the state figures out its paperwork: Arizona Public Health Association’s June 2026 analysis found predicted overdose deaths jumped 18 percent from 2024 to 2025, with deaths involving methamphetamine and other stimulants rising more than 15 percent even as the national overdose trend keeps improving. Arizona is one of the only states moving the wrong direction on overdose deaths in general, and it’s leaving one of the only tools built for its specific drug problem unfunded and unrequested.

Layer the geography on top and the gap gets almost absurd. A New York Times analysis of Maricopa County medical examiner data covering 2024 through March 2026 found drug deaths rise 40 percent once Phoenix’s daily high crosses 110 degrees, and nearly double past 115. Arizona State University researcher Raminta Daniulaityte told the Times that 80 to 90 percent of fentanyl users in the region also use methamphetamine — meaning the meth crisis isn’t separate from the opioid crisis Arizona has poured resources into. It’s tangled inside it, amplified by heat, and the treatment built specifically for the stimulant half of that equation is sitting on a table Arizona hasn’t walked up to.

What this looks like for someone actually trying to stop, and what a case manager can do about it this week

None of this is abstract if you’re the one white-knuckling a negative screen through a hundred-plus-degree afternoon. Methamphetamine and cocaine use disorder come with a specific, brutal symptom that makes early recovery feel impossible: anhedonia, the medical word for the flattening of pleasure — food, sex, sleep, connection, all of it dulled, because the drug has spent months or years training your brain’s reward circuitry to expect a much bigger spike than daily life can deliver. Contingency management works partly because it’s designed around exactly that problem. A modest reward for a negative screen doesn’t cure anhedonia, but it gives the reward system something immediate and real to register while the slower work of healing happens underneath it. The Veterans Health Administration data on this is now direct: a cohort study of 1,481 patients with stimulant use disorder who received contingency management, matched against 1,481 who didn’t, found the CM group was 41 percent less likely to die within a year. That’s not a craving-management footnote. That’s a mortality finding.

None of this is abstract if you’re the one white-knuckling a negative screen through a hundred-plus-degree afternoon.

If you’re a case manager or referral coordinator reading this, here’s the concrete move: this week, when you’re vetting or re-vetting a treatment partner for a client whose primary substance is meth or cocaine, ask them directly whether they run a contingency management program and, if so, how it’s funded. With no Arizona Medicaid waiver, any CM your client can access right now is almost certainly riding on a SAMHSA discretionary grant, like State Opioid Response funding, not standard billing — which means it can vanish when that specific grant cycle ends, sometimes mid-treatment. Ask for the grant name and the renewal date, not just a yes. Knowing which of your partners has real, funded CM versus which one just uses the phrase in a brochure is the difference between a referral that helps and one that quietly fails your client in month three.

Arizona already has one honest, working answer to its heat problem: Maricopa County distributed 1,700 naloxone-stocked heat relief kits this year, aimed at the fentanyl side of the crisis, and it’s real, it’s happening now, and it’s saving lives this month. But naloxone reverses an opioid overdose in progress. It does nothing for the person whose meth use is quietly wrecking their sleep, their job, and their thermoregulation three weeks before any 911 call. Arizona built the emergency response. It still hasn’t built the treatment. Every July until it does, that gap gets measured the same way it always has — in a medical examiner’s log, one hundred-plus-degree afternoon at a time.

Related: Stimulants coverage · Treatment & recovery · Arizona Watch