Skip to content

Model the ICHIA lawfully-residing option in Medicaid and CHIP - #9737

Merged
DTrim99 merged 9 commits into
mainfrom
medicaid-chip-lawfully-residing-option
Oct 5, 2026
Merged

DTrim99 merged 9 commits into
mainfrom
medicaid-chip-lawfully-residing-option

Conversation

@DTrim99

@DTrim99 DTrim99 commented Oct 1, 2026 •

Copy link
Copy Markdown
Collaborator

Summary

Closes #9736
Closes #9346

This adds the CHIPRA §214 option (often called "ICHIA") to Medicaid and CHIP. Through it, a state covers lawfully residing children and pregnant people without the five-year bar or the qualified-alien limit. The PR also gives CHIP its own immigration test. Until now, CHIP excluded only DACA recipients and undocumented immigrants. It now applies the five-year bar, the qualified-alien limit and the H.R.1 §71109 status limits, except where a state elected the option in both Medicaid and CHIP.

  • New parameters store each state's elections and dates:
    • gov.hhs.medicaid.eligibility.lawfully_residing.{immigration_statuses, child, child_age_limit, pregnant};
    • gov.hhs.chip.lawfully_residing.{child, pregnant}.
  • New is_medicaid_lawfully_residing_option_eligible (Person, YEAR) is ORed into is_medicaid_immigration_status_eligible.
  • New is_chip_immigration_status_eligible (Person, YEAR) replaces the inline ~(UNDOCUMENTED | DACA) checks in is_chip_eligible_child, is_chip_eligible_standard_pregnant_person, ca_cchip_eligible and ms_hmw_other_coverage_ineligible. FCEP is unchanged, because the unborn child is the beneficiary. The Healthier Mississippi Waiver check mirrors the CHIP child test, so an aged, blind or disabled child who fails CHIP on status now falls to the waiver category.

Law

42 USC 1396b(v)(4)(A), added by CHIPRA §214(a) and effective April 1, 2009 (§3(a)):

A State may elect (in a plan amendment under this subchapter) to provide medical assistance under this subchapter, notwithstanding sections 1611(a), 1612(b), 1613, and 1631 of title 8, to children and pregnant women who are lawfully residing in the United States [...] (i) Pregnant women — Women during pregnancy (and during the 60-day period beginning on the last day of the pregnancy). (ii) Children — Individuals under 21 years of age [...]

42 USC 1397gg(e)(1)(Q) applies the option to CHIP "but only if the State has elected to apply such paragraph with respect to such category of children or pregnant women under subchapter XIX".

From October 1, 2026, 1396b(v)(5) (H.R.1 §71109(a)) limits federal funding to citizens, permanent residents, Cuban and Haitian entrants and COFA migrants "except as provided in paragraphs (2) and (4)". 1397gg(e)(1)(R) (§71109(b)) applies that limit to CHIP.

CMS guidance:

  • SHO #10-006 defines "lawfully present", which includes qualified aliens and TPS holders.
  • SHO #12-002 excludes DACA recipients.
  • SHO #21-007 Q11 extends the 12-month postpartum period to the option's pregnant people.
  • SHO #26-001 confirms that §71109 leaves the option intact (p. 5) and that the five-year bar applies to CHIP (p. 6).

Behavior before and after

Person (in the five-year bar unless noted) Before After
LPR child or pregnant person, electing state, years_since_us_entry < 5 Fails Medicaid Medicaid immigration-eligible
Same person, years_since_us_entry ≥ 5 (the default) Eligible Eligible (unchanged)
TPS child or pregnant person, electing state (e.g. NY, TX child) Fails Medicaid (except CA) Medicaid immigration-eligible
Refugee, asylee, parolee child or pregnant person, electing state, 2027+ Fails Medicaid after the §71109 list narrows Stays Medicaid immigration-eligible
DACA or undocumented child Fails Fails (unchanged)
Recent-LPR or TPS child, separate-CHIP state without the election (AL, AZ, ID, IN, KS, MO, MS, SD, TN) CHIP-eligible Fails CHIP
Recent-LPR or TPS pregnant person, separate-CHIP pregnancy state without the election (MO only) Standard CHIP-eligible Fails standard CHIP; FCEP still covers the unborn child
Refugee or asylee child, separate-CHIP state without the election, 2027+ CHIP-eligible Fails CHIP (§71109)
Same people in a state that elected the option for CHIP CHIP-eligible CHIP-eligible (unchanged)

The option is age-based. A 19- or 20-year-old in an under-21 state qualifies in whatever group they otherwise meet. The pregnancy test uses is_pregnant_for_medicaid_nfc, which runs through each state's postpartum period.

State elections

The CMS list updated April 2, 2026 sets the current values, which match SHO #26-001's count of 39 states plus DC. Values before 2022 extrapolate from the KFF/CCF January 2022 survey, the earliest state-level source; it reconciles exactly with the current list minus the dated changes. "M-CHIP" states cover their option children in Medicaid-expansion CHIP, which the model treats as Medicaid, so their CHIP election is false.

Group States
Medicaid children under 19 AR CO FL GA HI IL KY LA MT NE NH NV NC OR RI TX UT VA WV WI
Medicaid children under 21 CA CT DE DC IA ME MD MA MI MN NJ NM NY OH PA SC VT WA
Medicaid pregnant AR CA CO CT DE DC GA HI KY ME MD MA MI MN NE NV NH NJ NM NY NC ND OH PA RI SC VT VA WA WV WI WY
Separate CHIP children AR CA CO CT DE FL GA IA LA MA MT NJ NV NY OR PA TX UT VA WA WV WI
Separate CHIP pregnant CO KY NJ RI VA WV
No election AL AK AZ ID IN KS MS MO OK SD TN
Dated change Date Source
RI pregnant (Medicaid and CHIP) 2022-10-01 SPA RI-22-0024 (election) and CHIP SPA RI-22-0025
NV pregnant 2023-01-01 SPA NV-22-0026
KY pregnant (Medicaid and CHIP) 2023-01-01 KFF Jan 2023 fn 7 ("in 2022"; first survey date showing it)
ND pregnant 2023-10-01 ND DHHS testimony
GA children, pregnant and CHIP children 2024-01-01 SPA GA-23-0002
NH children and pregnant 2024-01-01 SPA NH-24-0003
MI children (under 21) and pregnant 2024-08-01 MMP 24-20
IL CHIP children end 2022-07-01 KFF Jan 2022 "Y" and Jan 2023 "N/A (M-CHIP)"; matches the end of its separate CHIP in gov.hhs.chip.child.income_limit
KY CHIP children end 2022-07-01 Only the end of its separate CHIP in gov.hhs.chip.child.income_limit. KFF Jan 2023 still marks KY CHIP children "Y" (see Follow-ups)
ME, NC CHIP children end 2023-03-28, 2023-04-01 The end of their separate CHIP in gov.hhs.chip.child.income_limit

Formulas are YEAR-period, so a mid-year date takes effect in the next annual run (RI from 2023, ND from 2024, MI from 2025).

Oregon: CMS lists its option children under M-CHIP, and KFF reports the move on January 1, 2024. The model still covers Oregon children from 138% to 305% FPL through separate CHIP (gov.hhs.chip.child.income_limit). Its CHIP election therefore stays true, so those children keep coverage until the income limits reflect the move.

Tests

  • New is_medicaid_lawfully_residing_option_eligible.yaml (22 cases): CA under 21 and TX under 19, WY and ND pregnant-only, AL with no election, and TPS versus DACA and undocumented. It also covers postpartum (NJ at 90, 364 and 365 days; IL at 90 days without the pregnancy election), the effective dates (MI, GA, NH, ND, NV, KY, RI), the post-§71109 refugee cases and the age-limit boundaries. The postpartum cases use only 365-day states, so they do not depend on the stale 60-day values.
  • New is_chip_immigration_status_eligible.yaml (14 cases): the five-year bar at 4 and 5 years, the CHIP age limit of 19, CO versus TX pregnancy, DACA and undocumented, and refugees in 2026 and 2027. A 2027 AZ case covers asylees, withheld deportations, parolees and conditional entrants. It also tests the GA, KY, RI and IL dates and a one-override test that the CHIP option needs the matching Medicaid election.
  • Added cases:
    • is_medicaid_immigration_status_eligible.yaml (4);
    • emergency Medicaid integration.yaml (2): a TX LPR parent and child at 2 years, where the child gets Medicaid and the parent gets emergency Medicaid or none;
    • is_chip_eligible_child.yaml (6): AZ versus TX, refugees in 2025 and 2027, the GA dates;
    • is_chip_eligible_standard_pregnant_person.yaml (1): CO standard CHIP versus MO, where FCEP still applies;
    • ca_cchip_eligible.yaml (1): a recent-LPR child;
    • MS HMW ms_hmw_other_coverage_ineligible.yaml (1) and integration.yaml (1): a blind LPR child at 2 years fails CHIP and gets the waiver category and emergency Medicaid.

I ran the following locally, one process at a time. All 1,284 cases pass:

Suite Passed
gov/hhs/chip, gov/states/ca/dhcs/cchip, gov/hhs/medicaid/emergency_medicaid 235
gov/hhs/medicaid/eligibility (top-level files) 166
gov/hhs/medicaid/eligibility/categories 215
gov/hhs/medicaid/income, costs and the top-level Medicaid files 119
gov/aca (including the below-FPL exception) 286
gov/states/il/hfs/hbi, gov/states/or/oha/healthier_oregon, gov/hhs/basic_health_program, gov/states/wa/dshs/integration.yaml 150
Partners: healthcare/ plus the SSI and SNAP files that contain non-citizens 113

After the round-1 review fixes, I re-ran the two new files (21 and 14 cases), is_chip_eligible_child.yaml (22) and is_medicaid_immigration_status_eligible.yaml (18). All 75 pass. After the round-2 fixes, I re-ran the MS HMW tests (34), is_chip_immigration_status_eligible.yaml (14) and is_medicaid_lawfully_residing_option_eligible.yaml (21). All 69 pass. After the RI date fix, I re-ran is_medicaid_lawfully_residing_option_eligible.yaml (22), is_chip_immigration_status_eligible.yaml (14), is_chip_eligible_standard_pregnant_person.yaml (8) and is_medicaid_immigration_status_eligible.yaml (18). All 62 pass.

No existing expected value changed. The existing CHIP immigration cases are all in states that elected the option for CHIP (TX Case 15, CO Case 7, CA CCHIP) or use citizens. Existing LPR cases take the default of 5 years. Every other partner case is a citizen or undocumented, and the change cannot affect either group.

Microsimulation

Reading only the needed columns of the default dataset (populace_us_2024.h5, tag populace-us-2024-spm-20260909) with h5py:

  • the people are citizens (318.07M weighted), LPRs (11.03M), Cuban/Haitian entrants (0.73M), DACA recipients (0.30M) and undocumented immigrants (9.95M);
  • there are no TPS, refugee, asylee, parolee or conditional-entrant records;
  • there is no years_since_us_entry column.

Every LPR takes the default of 5 years and passes the bar. Cuban/Haitian entrants stay qualified and bar-exempt after §71109, and DACA recipients and undocumented immigrants are outside the option. So weighted Medicaid and CHIP enrollment does not change today, by construction. After PolicyEngine/microcosm#776 stores real entry years and #9732 removes the default, recent-LPR children and pregnant people will keep Medicaid in electing states. In separate-CHIP states without the election, they will lose CHIP.

Not in scope

Follow-ups

  • Oregon: move it to M-CHIP. Set its gov.hhs.chip.child.income_limit to -inf from 2024-01-01, cover those children through the Medicaid child limits, and flip gov.hhs.chip.lawfully_residing.child.OR to false in the same change. Doing only one of the two would drop recent-LPR and TPS children in Oregon.
  • Kentucky: check the 2022-07-01 end of its separate CHIP for children in gov.hhs.chip.child.income_limit, since KFF's January 2023 table still lists separate-CHIP children there.

Review history: local round 1 found 0 critical and 2 should items. The postpartum test fix is applied; the CHIP-side rulespec-us issue (42 USC 1397gg(e)(1)(Q)-(R)) is pending maintainer filing. Local round 2 found 0 critical and 1 should item, now fixed: the Mississippi waiver check still mirrored the old CHIP gate. Ziming requested changes (RI pregnancy date), fixed.

axiom: 42 USC 1396b(v)(4) encoded-correct (rulespec-us us/statutes/42/1396b/v.yaml) | 42 USC 1397gg(e)(1)(Q)-(R) TheAxiomFoundation/rulespec-us#1474 queued

🤖 Generated with Claude Code

https://claude.ai/code/session_01LqCGPgxDukji8Bhs57c5rH

DTrim99 and others added 3 commits October 1, 2026 10:07
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LqCGPgxDukji8Bhs57c5rH
- New is_medicaid_lawfully_residing_option_eligible: lawfully residing
  children (under 19 or 21 by state) and pregnant people in states that
  elected 42 USC 1396b(v)(4), ORed into Medicaid immigration eligibility.
- New is_chip_immigration_status_eligible replaces the inline
  undocumented/DACA check in CHIP: citizens, qualified aliens past the bar
  or bar-exempt (with the dated H.R.1 section 71109 list), and the CHIP
  lawfully-residing option (1397gg(e)(1)(Q)), which requires the Medicaid
  election.
- State election parameters dated from the CMS list, SPAs and bulletins.

Closes #9736
Closes #9346

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LqCGPgxDukji8Bhs57c5rH
@codecov

codecov Bot commented Oct 1, 2026 •

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 100.00%. Comparing base (909176a) to head (0e8578f).
⚠️ Report is 120 commits behind head on main.

Additional details and impacted files
@@            Coverage Diff            @@
##              main     #9737   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files            4         6    +2     
  Lines           76       128   +52     
  Branches         2         0    -2     
=========================================
+ Hits            76       128   +52     
Flag Coverage Δ
unittests 100.00% <100.00%> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

DTrim99 and others added 3 commits October 1, 2026 21:23
…arameter notes

- Move the postpartum boundary cases off California's stale 60-day value
  (Illinois at 90 days, New Jersey at 364/365 days).
- Note in the Medicaid status, bar-exempt and five-year-bar parameters that
  CHIP now reads them; correct the Kentucky CHIP end-date citation.
- Add a 2027 CHIP case for the other post-H.R.1 statuses in a non-elected
  separate-CHIP state.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LqCGPgxDukji8Bhs57c5rH
…ssissippi HMW

- ms_hmw_other_coverage_ineligible now reads is_chip_immigration_status_eligible
  instead of mirroring the old undocumented/DACA check, so a child who fails
  the CHIP immigration test is not treated as CHIP-covered for HMW.
- Wording fixes in the lawfully-residing status parameter and a test name.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LqCGPgxDukji8Bhs57c5rH

@hua7450 hua7450 left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Requesting changes to correct Rhode Island's pregnancy-election effective date in both Medicaid and CHIP, as detailed in the inline comment. CMS approved the elections effective October 1, 2022, so they should apply in the model's 2023 annual calculation.

All 140 cases in the nine changed YAML test files passed locally. An independent 2023 Rhode Island case fails for pregnant TPS holders and recent permanent residents; a control overriding only the two election parameters passes, including the derived Medicaid eligibility and CHIP immigration outputs.

DTrim99 and others added 2 commits October 2, 2026 14:02
Medicaid SPA RI-22-0024 and CHIP SPA RI-22-0025 approve the lawfully
residing pregnant-individual elections effective October 1, 2022; update
both parameter entries, references and the 2022/2023 RI test cases.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LqCGPgxDukji8Bhs57c5rH
@DTrim99

DTrim99 commented Oct 2, 2026

Copy link
Copy Markdown
Collaborator Author

Thanks, Ziming. Rhode Island's pregnancy elections now start on October 1, 2022, in both Medicaid and CHIP.

Parameters

  • gov/hhs/medicaid/eligibility/lawfully_residing/pregnant.yaml:98-101: RI now changes on 2022-10-01: true (it was 2024-01-01). The inferred-date comment now cites SPA RI-22-0024. The references at lines 136-139 cite the SPA's PDF pages 4 and 11.
  • gov/hhs/chip/lawfully_residing/pregnant.yaml:90-93: the same change, with the comment citing CHIP SPA RI-22-0025. The reference at lines 134-135 cites the approval letter, PDF page 2.
  • I removed the KFF May 2024 reference from both files. It only bracketed the old inferred RI date.

Tests

  • is_medicaid_lawfully_residing_option_eligible.yaml:
    • Case 21 (line 462) now runs your 2023 case. Two pregnant 30-year-olds in RI at 150% FPL, one with TPS and one an LPR two years after entry, expect true, true for the option, is_medicaid_immigration_status_eligible and is_medicaid_eligible.
    • New Case 22 (line 489) keeps the boundary: 2022 is still false under the January-1 annual convention.
    • Case 20 (2024, true) was only renamed.
  • is_chip_immigration_status_eligible.yaml: Case 9 (line 183) now expects true for RI in 2023, and I added the two-year LPR. Case 8 (2022, RI false) is unchanged.

I ran four files one at a time, and all pass: the Medicaid option file (22), CHIP immigration (14), is_chip_eligible_standard_pregnant_person (8) and is_medicaid_immigration_status_eligible (18). No other expected value changed.

SPA pages I checked

  • RI-22-0024 (13 pages, approved April 19, 2023):
    • PDF page 4, the Submission Summary, lists "Citizenship and Non-Citizen Eligibility" with a proposed effective date of 10/1/2022. It supersedes 13-022.
    • PDF page 11, "C. Coverage of Lawfully Residing Individuals", shows "Yes" selected for the 1903(v)(4) option and "1. Pregnant women" checked. Children under 19 are also selected, which matches the existing RI child entry. The page header gives an effective date of 10/1/2022.
    • The approval letter on page 2 gives October 01, 2022, but it describes only the postpartum election. So pages 4 and 11 are the right citations.
  • RI-22-0025 (5 pages): the letter's first page, PDF page 2, says the SPA "has an effective date of October 1, 2022". It calls the SPA a companion to RI-22-0024 and says it "expands CHIP coverage to otherwise eligible lawfully residing pregnant individuals under section 214" of CHIPRA.

@DTrim99
DTrim99 requested a review from hua7450 October 2, 2026 18:02

@hua7450 hua7450 left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The prior Rhode Island pregnancy-election finding is resolved in both Medicaid and CHIP. All CI checks passed on reviewed commit ccfdf69. Approving following the local re-review.

@DTrim99
DTrim99 merged commit 9d6c274 into main Oct 5, 2026
35 checks passed
@DTrim99
DTrim99 deleted the medicaid-chip-lawfully-residing-option branch October 5, 2026 13:05
DTrim99 added a commit that referenced this pull request Oct 26, 2026
…rging main

Main added two direct readers of years_since_us_entry (CHIP immigration
eligibility from #9737 and New Hampshire adult cash assistance) and two
LIHEAP programs (Nebraska, Oklahoma) that read SNAP immigration eligibility,
including the qualified-alien waiting period.

- is_chip_immigration_status_eligible Case 7: the Arizona permanent resident
  checks the post-71109 status list, so it now states five years.
- Case 14: the comment no longer relies on the default of 5; the parolee and
  conditional entrant state five years so only the status limit applies.
- New Case 2b locks the default of 0 for CHIP (inside the bar).
- ne_liheap_household_size Case 2: the permanent resident couple checks that
  permanent residents count, so both state five years.
- years_since_us_entry documentation names the programs that apply the bar.

No expected value changed.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LqCGPgxDukji8Bhs57c5rH
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

2 participants