You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Add broad Medicare enrollment targets to the healthcare fiscal surface #231
The current healthcare target surface includes Medicare Part B premium income, and #169 tracks a payer-count shortfall for that amount target. It does not yet include broader Medicare enrollment targets.
That leaves several important Medicare surfaces absent from release-over-release healthcare diagnostics: Part A enrollment, Part B enrollment, Medicare Advantage, Part D, dual eligibles, and LIS.
Goal
Add a Medicare enrollment target cluster, or explicitly document which Medicare enrollment surfaces are not yet available to Populace.
Suggested implementation
Inventory CMS Medicare enrollment source artifacts available to Ledger/Populace.
Add target roles for supported enrollment concepts, especially:
Medicare Part A enrollment
Medicare Part B enrollment
Medicare Advantage enrollment
Medicare Part D enrollment
dual eligibles
LIS recipients
Prefer national and state targets where source geography supports them.
Keep enrollment targets separate from the existing Part B premium-income target so diagnostics can identify payer-count vs amount-per-payer problems.
Acceptance criteria
Release artifacts expose Medicare enrollment targets where source support exists.
Metadata names the target role and source concept for each Medicare enrollment surface.
Tests cover target mapping and profile coverage for the Medicare enrollment cluster.
That PR sets the model's Medicare value per enrollee to CMS's NHE Table 17 Medicare spending per enrollee ($20,096 in 2027, $30,129 in 2034). On the default dataset (populace_us_2024.h5@populace-us-2024-spm-20260909), weighted medicare_enrolled is then the only reason modeled Medicare spending differs from CMS's:
Year
Modeled enrollees
CMS enrollees (NHE Table 17, Part A and/or B person-years, 50 states and DC)
Ratio
Modeled medicare_gross_cost
NHE Medicare spending
Ratio
2027
58.80M
70.8M
83.1%
$1,181.6B
$1,422.7B
83.1%
2034
60.27M
78.1M
77.2%
$1,815.8B
$2,354.3B
77.1%
Modeled enrollment grows 2.5% from 2027 to 2034; CMS projects 10.3%. A total Medicare enrollment target (Part A and/or B), with its projected path, would close both gaps.
Activity
Re-pin the UK target surfaces to Chronicle 6fb700e and repair the con…
Re-pin the UK target surfaces to Chronicle 6fb700e and repair the con…
MaxGhenis commentedon Oct 12, 2026
A measured data point for this issue, from the microsim runs behind PolicyEngine/policyengine-us#10107.
That PR sets the model's Medicare value per enrollee to CMS's NHE Table 17 Medicare spending per enrollee ($20,096 in 2027, $30,129 in 2034). On the default dataset (
populace_us_2024.h5@populace-us-2024-spm-20260909), weightedmedicare_enrolledis then the only reason modeled Medicare spending differs from CMS's:medicare_gross_costModeled enrollment grows 2.5% from 2027 to 2034; CMS projects 10.3%. A total Medicare enrollment target (Part A and/or B), with its projected path, would close both gaps.
Source: NHE projections tables 2025-2034, https://www.cms.gov/files/zip/nhe-projections-tables.zip (Table 17, rows "Medicare" under spending, enrollment and spending per enrollee). Run details are in the PR description.
🤖 Generated with Claude Code