Population model · ACOs · health systems · MA plans
What deprescribing returns on your population
Adjust the assumptions to your book, then view the return the way your organization is measured: Star measures, total cost of care, or utilization. Defaults are deliberately conservative, and every assumption is verified against your claims before we quote impact.
Your population
Any benzo/z-drug use, high anticholinergic burden, or CNS-active polypharmacy, deduplicated — broad definition. 25% default; strict PQA definitions run 10–20%. Verified in your claims.
15% default. Best published benchmark: 26% taper initiation in a pre-curated primary-care cohort (JAGS 2024); cold claims-flagged outreach converts lower.
43% default — D-PRESCRIBE’s six-month sedative-hypnotic rate vs. 12% usual care (JAMA 2018). Dollar math below uses the net program effect.
Where the savings number comes from
We model $250–$1,000 per program-attributable discontinuation, per year — anchored to the observed $1,163/yr all-cause cost difference in anticholinergic-polypharmacy-positive Medicare members (Campbell et al., Drugs & Aging 2021), then discounted because that figure is an association, not a measured post-taper saving. The 12% who quit under usual care are netted out before any dollar is claimed.
On a target med
12,500
Enrolled
1,875
Off med, year 1
~806
Modeled avoided cost · program-attributable, year one
$145K–$581K
~806 members off a high-risk medication in year one — and the modeled delta recurs every year they stay off, so the impact compounds as cohorts accumulate.
~5
injurious falls avoided, year one
~4
fall-related ED visits avoided
~2
fall admissions avoided, at ~$19K each
Event counts apply the benzodiazepine-attributable fall fraction (28% — adults 80+, PAQUID cohort, Drugs & Aging 2008) with a phased-reversal discount: only ~40% of attributable risk is credited in year one, because fall risk reverses over months, not instantly. Both events and dollars net out the 12% usual-care discontinuation rate.
What your plan gets · Stars (COB & Poly-ACH)
~1.6 pts off your measure rate
COB and Poly-ACH enter the Star Ratings with measurement year 2025 — the 2027 Stars; Poly-CNS remains a display measure. The scored measures are computed from Prescription Drug Event (PDE) data, so the only way to move them is to actually move members off the drugs. And on analogous scored Part D measures, adjacent star thresholds sit just 2–4 points apart: a 1–2 point rate reduction can move a large plan a full star, and the plan keeps 100% of that value.
Timing: a member who accrues 30+ cumulative days of qualifying exposure counts against the measure for that entire measurement year, so completed tapers move the following year’s rate — and early starts keep members from ever crossing the threshold. COB/Poly-ACH have no CMS-published cut points yet; the one-star illustration is benchmarked to analogous scored measures (SUPD, PDC adherence).