Arizona's Health Department Says Overdoses Are Down More Than Half. The CDC Says Arizona Is One of Three States Where They're Not.

Two Arizona overdose statistics have been circulating within weeks of each other this summer, and they cannot both be describing the same reality. In May, CDC’s National Center for Health Statistics reported that Arizona was one of only three states — with New Mexico and Colorado — where overdose deaths rose 10% or more in 2025 compared to 2024, even as the national rate fell roughly 14%. In June, the Arizona Department of Health Services told local reporters that opioid overdoses in Maricopa County had dropped by more than 2,300 — over half — with Pima and Pinal counties reporting similarly steep declines.

A state cannot meaningfully manage an overdose crisis it cannot describe consistently to its own residents, and right now Arizona is describing it two different ways at once.

The honest answer is probably not that either agency is wrong — it’s that they’re very likely measuring two different things and calling them by the same name. CDC’s provisional figure is a 12-month count of confirmed and estimated overdose deaths, compiled from death certificates that take months to finalize and get explicitly labeled “incomplete and subject to change” in the agency’s own release. ADHS’s number, drawn from more real-time county-level surveillance, more plausibly reflects reported opioid overdose events — which can include nonfatal overdoses reversed by naloxone — over a shorter or differently bounded window. A steep drop in reported nonfatal events and a rise in confirmed deaths aren’t even necessarily in tension: it’s possible to see fewer overdoses being caught in real time while the ones that do happen have gotten more lethal, given that Arizona’s own synthetic-opioid death rate is reportedly up 33% and cocaine-related deaths up nearly 70% year over year.

That’s a plausible reconciliation. It’s also not one either agency has offered publicly, and until one does, both numbers get quoted at face value by whoever finds them useful — county officials citing the drop, public health researchers citing the rise, neither obligated to explain why their number doesn’t match the other. For a state overseeing $1.2 billion in incoming opioid settlement money and being asked, correctly, to fund “programs that can show measurable results,” a first step that costs nothing is publishing what “results” is actually counting.