Arizona’s Overdose Deaths Are Still Rising. The Slowdown Is the Story.

New Mexico’s predicted overdose deaths rose 27.1 percent in the twelve months ending February 2026 — up from 816 the year before to 1,037. Colorado rose 14.2 percent. South Dakota, North Dakota, Minnesota, Montana: all rising, all part of what Filter’s Kastalia Medrano calls a western corridor bucking the national trend, where overdose deaths are still falling nationally but at a much slower pace than a year ago — 12.1 percent down, versus a 26-percent drop in 2024.

Arizona belongs on that list too. Predicted deaths climbed from 2,695 to 2,914. But the trajectory is doing something its neighbors’ isn’t: in the April 2026 CDC report, Arizona’s year-over-year increase stood at 21 percent. By July, it had dropped to 8.2 percent. New Mexico moved the opposite direction over the same stretch, from 16 percent to 27.1. Same regional headwinds — fentanyl supply, rural access gaps, a drying-up of pandemic-era harm reduction funding — and Arizona is the one state in the corridor visibly pulling out of the dive.

Say the honest version first: rate-of-increase improving is not deaths falling. Arizona buried more people to overdose this year than last. Nobody should read “8.2 percent” as good news. It’s less bad news, three data points running in the right direction, which is a real thing but a fragile one — CDC provisional estimates get revised, and one soft quarter doesn’t make a trend.

That said, I don’t think the divergence is coincidence, and I’m not going to hide behind “too early to say” just because the causal chain isn’t proven. Arizona has been doing something concrete that New Mexico, as far as public reporting shows, hasn’t matched at the same scale. Maricopa County has pushed out 1,700 naloxone-stocked heat relief kits this year, built specifically around the fact that extreme heat raises fatal overdose risk for people who use opioids — and county health official Carmen Batista says the kits have already produced at least two confirmed reversals, including a parent who saved their own child. Meanwhile AHCCCS opened its first Rural Health Transformation grant, an “Opioid Antagonist Distribution Initiative” aimed at rural counties, applications due August 7 — money that hasn’t even landed yet, so it can’t explain the deceleration that’s already shown up in the data.

Two documented naloxone saves and one grant application deadline are not, on their own, an explanation for a 900-death-a-year state’s improving trajectory. But they’re the kind of intervention that would produce exactly this signature: fewer deaths at the margin, not fewer overdoses. Naloxone doesn’t stop anyone from using. It stops the ambulance call from becoming a funeral. If Arizona’s numbers keep bending while New Mexico’s keep climbing, that’s the mechanism I’d bet on first — and the volunteer who used a kit from a heat-relief table to save their own son didn’t need a CDC report to know it worked.