Why this exists

Most overdose data an agency acts on describes a period that has ended. Provisional national counts run months behind; final counts run longer. Meanwhile the local picture — which neighborhoods, which substances, which times, which services people did or did not reach — is scattered across emergency medical services, emergency departments, medical examiners, harm-reduction organizations and law enforcement, none of which sees the others’ data in useful time.

The consequence is that response allocation is done against last year’s pattern. Naloxone goes where overdoses were, outreach goes where outreach went, and the current cluster is invisible until it is historical.

How it works

• Event-level intake, aggregate output. The registry ingests overdose events and publishes patterns — location, substance where known, time, outcome, service contact — at the level a response can act on.
• Multiple contributors, one picture. The value is entirely in the integration; each contributor already has its own fragment.
• Timeliness as the design constraint. A registry that is complete and three months late has failed at the only thing it was built for.
• Linked to emerging threat analysis, so that a change in the pattern triggers a warning rather than a report.