The Line Items They'd Kill Are the Ones Funding the Human Work

On the morning of January 13, 2026, substance use disorder programs across the country received emails from the Substance Abuse and Mental Health Services Administration. Six lines. No preamble. The subject line, in many cases, was some variant of “Notice of Award Termination.” By noon, the calls were coming into the National Council for Mental Wellbeing’s member hotline faster than staff could answer them. Syringe service programs in rural Appalachia. Peer recovery organizations in Albuquerque. A workforce development training center in Louisville that had been running on the same federal grant for nine years. All terminated, effective immediately.

Twenty-four hours later, HHS reversed course. The terminations — affecting more than 2,800 grants totaling roughly $2 billion — were rescinded after bipartisan congressional outrage, a wave of statements from advocacy organizations, and what one program director in West Virginia described as a “collective panic attack” across the behavioral health sector. The grants were restored. The emails explaining the reversal were also about six lines long.

But what the administration tried to do in a day in January is now, six months later, being attempted by statute. The White House FY2026 budget proposal would replace three of SAMHSA’s existing grant programs with a single Behavioral Health Innovation Block Grant funded at $4 billion. That’s down from $4.2 billion in current combined spending — a 5% nominal cut. And that 5% figure, which has circulated widely in media coverage and congressional testimony alike, is the wrong number to focus on.

What a block grant consolidation actually does

The mechanics matter here. Under the current system, SAMHSA distributes money through a mix of formula block grants and discretionary grant programs. Formula block grants — like the Community Mental Health Services Block Grant and the Substance Use Prevention, Treatment, and Recovery Support Services Block Grant — pass through state governments. They’re governed by statute, require maintenance-of-effort provisions, and carry categorical requirements that tell states what they have to spend the money on. States dislike these requirements. They call them inflexible. From a treatment infrastructure standpoint, those requirements are the point.

Discretionary grants are different. These are the programs SAMHSA funds directly: grants to individual organizations, coalitions, and tribal governments for specific, narrowly defined work. They operate outside the block grant formula entirely, which means SAMHSA can direct them toward activities that voluntary state spending never would. Harm reduction. Peer recovery workforce development. Overdose surveillance infrastructure. Youth prevention in communities that can’t sustain a full treatment program. The $813 million in “grants of regional and national significance” — what SAMHSA calls the line items that fund this kind of work — would be eliminated entirely under the proposal.

Entirely. Not reduced. Not consolidated. Gone.

The Behavioral Health Innovation Block Grant would replace the existing framework with a single stream distributed by formula to states with broader “flexibility” in how they use the money. The accompanying budget justification, according to analysis published in Health Affairs Forefront, frames this as streamlining. It cites the consolidation of three grant streams into one as an administrative simplification. It describes the new block grant as an opportunity for states to direct resources toward locally determined priorities.

What it does not say, but what the historical record on block grants is clear about, is this: when states get flexibility, they tend to use it to fund the politically popular and clinically familiar. Inpatient detox. Residential treatment. Things that look like treatment to legislators and constituents who don’t understand addiction science. What they don’t fund — given the choice — is syringe services. Peer recovery workers who themselves have a history of use. Data collection about overdose patterns in zip codes that overlap with their political donors’ zip codes. These are the categorical mandates that the current system forces. Without them, there is no force.

What they don’t fund — given the choice — is syringe services.

The $813 million nobody is talking about

The headline number in budget coverage has been $4 billion, compared to $4.2 billion — a $200 million trim. The number that tells you what the proposal actually does is $813 million: the total 2025 appropriation for grants of regional and national significance, which the proposal eliminates in full.

This $813 million is where SAMHSA’s programmatic work lives. It funded 17 categorical programs in fiscal year 2025, covering areas including:

Syringe service programs and harm reduction supplies, including fentanyl test strips — the same test strips the administration briefly made ineligible for federal funding in early 2025 before legal pressure forced a reversal. The workforce development pipeline: training for peer recovery specialists, counselors, addiction medicine physicians, and community health workers. Data infrastructure: the drug surveillance and overdose monitoring systems that tell us which drugs are killing people, where, and in what combinations — without which any allocation decision is educated guessing. Youth prevention programs and early intervention. Recovery support services for people leaving incarceration. Tribal behavioral health infrastructure, which rarely attracts voluntary state funding and rarely should have to.

None of this is in the Behavioral Health Innovation Block Grant. The $4 billion figure covers only the formula components of the existing structure — the part that passes through states. The $813 million was never in that formula. Collapsing the three grant streams into one block grant doesn’t include it; it erases it.

Maia Szalavitz, writing in Filter in February 2026, made the most direct version of the argument: “Every time the federal government replaces categorical addiction funding with block grants to states, harm reduction loses. Not because harm reduction is ineffective — the evidence is overwhelming that it is — but because harm reduction is unpopular in the states where it’s needed most.”

That’s the record. In the 1980s, block grant consolidation under Reagan reduced the reach of community mental health infrastructure that took a decade to rebuild. The Ryan White CARE Act exists specifically because states couldn’t be trusted to direct AIDS-related funding toward the communities most affected. The principle isn’t that states are incompetent; it’s that politically vulnerable populations get deprioritized when the mandates go away.

What the January terminations revealed

The administration’s January 2026 exercise was, in retrospect, a reveal. Not a policy mistake. A preview.

When SAMHSA sent those termination notices, the programs that surfaced in the immediate aftermath as most endangered shared a profile: discretionary grants to harm reduction organizations, peer recovery support programs, and regional training coalitions. These are not the formula block grant recipients that get to call their congressional representatives and invoke statutory authority. They’re organizations that exist because SAMHSA chose to fund the work directly, because states wouldn’t.

The reversal came fast because the political cost was too high, too visible, too sudden. A hundred programs talking to local journalists in the same week is not a manageable news cycle. A budget line item eliminated in a March appropriations bill is a different kind of decision — quieter, distributed, harder to attribute to a single moment.

Behavioral Health Business reported that many programs, even after the reversal, had already begun contingency planning: notifying staff, pausing hiring, reaching out to state health departments about potential bridge funding. The reversal restored the grants, but it didn’t restore certainty. Several harm reduction coordinators interviewed by regional outlets in subsequent months said they were operating with a parallel “shutdown plan” that they updated after every piece of congressional budget news.

This is the real cost of the January episode that gets lost in the “grants restored” framing: the planning tax. Every hour a program director spends building a contingency budget is an hour not spent running the program. Every staff meeting about “what we do if the funding disappears” is a staff meeting not about how to reach more people. The uncertainty is itself a harm.

Every hour a program director spends building a contingency budget is an hour not spent running the program.

The argument for block grants — and why it doesn’t hold here

The case for block grant consolidation is not nothing. State governments often do know more about local conditions than federal program officers in Rockville, Maryland. Categorical mandates can create perverse incentives — a grant that requires funding syringe programs regardless of local epidemiology is a blunt instrument in communities where opioids aren’t the primary driver of overdose deaths. Flexibility has value. Administrative simplification has value.

The problem is the specific combination of what’s in and what’s out of this proposal.

The $4 billion block grant retains the formula distributions that states already know how to absorb. It eliminates the $813 million that went to things states were explicitly not funding on their own. If the argument for consolidation is “states know best,” it should apply equally to the formula programs being preserved. The selection — keep the formula, cut the discretionary — isn’t a neutral administrative simplification. It’s a funding philosophy that says peer recovery workers, harm reduction infrastructure, and overdose surveillance are optional features of the behavioral health system.

They are not optional features. They are the part of the system that keeps people alive long enough to access treatment.

What this means in Arizona — and for anyone trying to build in this space

Arizona’s position is complicated. The state receives opioid settlement funds that could theoretically buffer some of what federal categorical grants have paid for — Maricopa County alone has begun disbursing settlement contracts for harm reduction and treatment navigation work. But settlement funds are time-limited, encumbered by oversight requirements that don’t always match program needs, and focused by their legal terms on opioid-specific interventions. They cannot replace a workforce development infrastructure that spanned all substance use disorders. They cannot replace surveillance funding. They do not cover peer recovery work for alcohol use disorder, or stimulant use disorder, or the polysubstance cases that now represent the majority of the overdose caseload.

The strategic question for treatment access platforms like Rize Recovery — and for any organization trying to connect people to services — is what the map looks like when the $813 million is gone. The answer isn’t merely that there are fewer services; it’s that the remaining services become less findable, less navigable, and more concentrated in the clinical tier that has always been harder for the uninsured and underinsured to access. The peer recovery workforce that was the connective tissue between the emergency department and the treatment program is the first to be defunded. The syringe service van that was the first point of contact for someone not yet ready for treatment is next.

The 5% cut doesn’t begin to describe what is actually proposed. What is proposed is the elimination of the federal mandate that the behavioral health system do the unglamorous, underfunded, evidence-based work that states have historically chosen not to do on their own.

That work is not decorative. For people in acute need of help, it is often the only path in.

---

Rize Recovery maintains an up-to-date map of treatment and harm reduction resources in Arizona. Find services near you or explore our policy and funding coverage to stay current on how federal changes affect local access.