GUBMENTPlain talk · policy frontier
Filings / Drugs / Sources / §2 Findings: Baseline
GBMT-8 · Research record · No. 8

§2 Findings: Baseline

drugs/research/ws02-findings.md
This is a working research document from the drugs filing, published as written — including the parts later corrected. It is the underlying record for Whitepaper No. 8, not a summary of it.

Date: 2026-08-03. Extends Phase 0 anchors 2 (treatment gap) and 3 (budget split); does not re-derive them.

1. Treatment capacity infrastructure

SAMHSA's N-SUMHSS (facility census) covered 20,681 eligible SUD/mental-health facilities in 2023, rising to 21,205 in 2024 — but this combines substance-use and mental-health facilities together. A KFF analysis isolating the 2022 wave found ~14,700 facilities specifically providing substance use treatment, with 34 beds per 100,000 population nationally (range 8–79 by state) and 71% accepting Medicaid (state range: 13 states below 70%, 16 states at 90%+).

Access deserts are real and quantified. HHS OIG (Sept. 2024): only ~22% of US counties (and just 6.9% of rural counties) had a Medicare-enrolled opioid treatment program in 2022. 30.1% of rural counties vs. 10.4% of urban counties had zero DEA-waivered buprenorphine prescribers as of end-2022. This is the geographic-access analog to §4's regulatory-cage finding — even where MOUD access exists on paper, reimbursement and provider-directory gaps mean many providers "did not treat any Medicare or Medicaid enrollees" in practice.

2. TEDS admissions

Latest fully-extracted TEDS data (2022): ~1.5M admissions, primary substance at intake alcohol 35.5%, heroin 16.9%, methamphetamine 13.9%, other opiates/synthetics 11.7%. TEDS is admission-based (not client-based — the same person generates multiple records across episodes) and reports administrative status-at-discharge only (length of stay, reason for leaving, discharge setting) — not clinical outcomes (abstinence, relapse, mortality post-discharge). This confirms and sharpens Phase 0's finding: the US has no national instrument measuring whether treatment actually worked, only whether someone was admitted and how they left.

3. Alcohol and tobacco comparators (protocol anchor row 10 — now verified)

This is a genuinely important framing point for §10/§11: the two legal drugs the protocol scoped in kill several multiples more people per year than the entire illegal-drug overdose crisis, using well-established, low-intensity policy levers (taxation, age-gating, marketing restriction — the tobacco playbook already in the seed architecture list). Any "drugs" policy conversation that stays confined to illegal substances is, by simple body count, looking at the smaller problem.

4. SUD prevalence detail — a genuine data gap

Could not extract current (2024) age-stratified SUD percentages — a tool/access limitation on SAMHSA's PDF/HTML report formats, not a confirmed data-unavailability. Directionally: SUD prevalence declined among adults 26+ from 2021–2024 but showed no significant change among 12–17 or 18–25 year-olds; 18–25-year-olds have the highest use rates across nearly every category while being least likely to receive treatment. The best available race/ethnicity breakdown is five-to-nine years stale (pooled 2015–2019 NSDUH: American Indian/Alaska Native highest at 12.2%, then multiracial 10.8%, White 8.2%, Black 7.6%, Hispanic 7.5%, Asian 4.3%) — flagged as a real gap for full execution rather than papered over with a current-sounding restatement of old data.

5. Total US SUD treatment spending — unresolved, not just unverified

This is the most consequential finding of the pass. Two credible source lineages disagree by roughly 3x for ostensibly the same quantity:

No reconciliation between these lineages was found in the literature. This should be presented in the whitepaper as an open methodological dispute — likely driven by different rules about what counts as "SUD treatment" (settings included, MH/SUD attribution, criminal-justice-funded treatment) — not resolved by picking whichever number is more convenient for a given argument.

Implications for §10/§11

Confidence: Moderate

Strong on facility/access-desert data (primary federal sources, cross-corroborated). Weak on current age-stratified prevalence and the race/ethnicity breakdown (stale, single-root). The spending discrepancy is reported as a genuine unresolved dispute, not a confidence-weak guess.

← All Drugs research documents Sources digest Read the whitepaper