Overdose Forecast

Built for Sara Carter · Director, White House Office of National Drug Control Policy

Forecast where overdose surges are likely in the next two weeks from leading indicators — so resources move before the spike, not after.

✎ Tried it? Share pilot feedback →

← Forecast board

Bernalillo Co., NM

aggregate region · pop ~676k · 2-week forecast to ~Jul 10

⚑ Surge likely

Forward risk

band 57–77 · 67% conf

Top driver

Strip +

61% of the forecast

Data confidence

2/4 feeds

complete 50% · fit 100% · agree 63%

Sparse data. 2 feed(s) are offline for this region, so the model redistributes their weight and widens the confidence band (57–77). The forecast is shown without false precision — treat it as a signal to confirm, not a settled number.

Leading indicators · last 3 weeks, projected +2w

EMS naloxone administrations

now 5.1/100k·wk · trend +0.45/wk · → 6.0 projected

Jun 12Jun 19Jun 26+2w

range 0–6

pressure 50

×0.53 = 27

Test-strip positivitydriver

now 61% · trend +6.50/wk · → 74.0 projected

Jun 12Jun 19Jun 26+2w

range 0–74

pressure 87

×0.47 = 41

ED overdose visitsfeed offline

No aggregate feed reporting for this region this cycle — excluded; its weight was redistributed.

Recent suspected lossesfeed offline

No aggregate feed reporting for this region this cycle — excluded; its weight was redistributed.

How Bernalillo Co.'s 67 is assembled

EMS naloxone  pressure  50 × eff.weight 0.53 = 27
Strip +       pressure  87 × eff.weight 0.47 = 41
forward risk = 67 → surge likely · confidence 67% → band ±10 (57–77)

Aggregate public-health indicators only — this region record holds no individual or person-level field of any kind. Forecast is a decision-support signal for pre-positioning, not a medical determination. Not medical advice.

Showcase demo for the America's Future 80th Anniversary Summit. Showcase · best home: Nonprofit / public-safety. Seeded with realistic sample data; runs payment-dark (no billing). Not a production system.