Date: 2026-08-06. Status: A late evidence check against claims already in the research record. It does not re-score the board or alter the live site; the scorecard needs a fresh reconciliation before any changed cell is used.
Contingency management: efficacy is not a completed legal pathway
A 2026 systematic review/meta-analysis of 26 trials (2,356 people) found longer continuous concurrent abstinence and more dual-negative samples from CM, but no overall retention effect. The trials are largely cocaine/heroin-era and do not establish fentanyl-era mortality.
SAMHSA's January 2025 advisory permits eligible grant recipients up to $750 per person per year. It says the existing safe-harbor amount was $605 in 2025 and does not create a CM-specific safe harbor or nationwide Medicaid/commercial reimbursement rule. The appropriate architecture label is therefore CM reimbursement and implementation expansion, not “CM legalization/safe-harbor expansion.” A broad change remains prospective OIG rulemaking plus payer implementation.
MOUD and harm-reduction components must not be bundled
- An urban-ED RCT found 30-day treatment engagement of 78% for ED-initiated buprenorphine versus 37% referral and 45% brief intervention. It did not show a mortality effect.
- A BC linked-administrative cohort associated in-custody OAT with lower 30-day nonfatal overdose after release (aHR 0.55), but remains observational and Canadian.
- Two evaluations of waiver removal find that expanded legal authority alone did not consistently increase actual prescribers or patients. This supports the existing legal-change-is-insufficient conclusion, not a claim that each MOUD delivery component has the same evidence base.
- A Massachusetts municipality interrupted time series supports naloxone as one contributor to lower fatal overdose (annualized rate ratio 0.84). A 2025 supervised-consumption review finds on-site safety and linkage benefits but indeterminate population-level mortality evidence. Naloxone and supervised consumption must be scored separately.
Attribution and decriminalisation boundaries
The Science supply-shock hypothesis is a plausible inference, but a 2026 geographic reexamination finds purity alone insufficient and flags time-series/autocorrelation concerns. The national decline should therefore be described as multi-cause and unresolved, not as a one-cause supply-shock result.
For Oregon M110, a 2024 matrix-completion synthetic-control study found no detected mortality association after modeling fentanyl's spread. That supports a limited mortality-null statement; it does not establish the success of the complete decriminalisation-plus-treatment package or resolve public-order and delivery outcomes.
Required scorecard follow-up
Before another board pass: split A2 into ED/community/telehealth and correctional components; A3 into treatment efficacy and legal/payer pathway; and A5 into naloxone, drug checking, and supervised consumption. Update the root-cause language before using supply-side evidence to justify an A8 mortality cell.