[hua7450]

Fixes #10362

What and why

is_chip_eligible_standard_pregnant_person required is_pregnant, so a person in the separate CHIP pathway for "targeted low-income pregnant women" lost CHIP at birth. Federal law keeps them eligible after birth:

• 42 USC 1397ll(d)(2)(A) covers them through the end of the month in which the 60-day postpartum period ends. Where the state applies the 12-month option of 42 USC 1396a(e)(16), coverage runs through the end of the month in which the 12-month period ends. The model approximates the month-end rule with day counts; Medicaid postpartum coverage should run to the end of the month in which the 60-day or 12-month period ends #10463 tracks it.
• 42 USC 1397gg(e)(1)(K) requires the 12-month coverage in separate CHIP once a state elects it in Medicaid.
• CMS SHO 21-007 (pages 4 and 7) applies the extension to targeted low-income pregnant women in a separate CHIP. Page 17 (Q11) extends it to lawfully residing pregnant people under the CHIPRA section 214 option. Pages 4 and 7 exclude unborn-child (FCEP) coverage.

This PR changes three variables:

• is_chip_eligible_standard_pregnant_person now uses is_pregnant_for_medicaid_nfc, the Medicaid pregnant-or-postpartum test. That test reads the per-state gov.hhs.medicaid.eligibility.categories.pregnant.postpartum_coverage parameter, so CHIP follows each state's Medicaid postpartum period without a new CHIP parameter.

• The lawfully residing pregnancy branch of is_chip_immigration_status_eligible uses the same test. The Medicaid version, is_medicaid_lawfully_residing_option_eligible, already does.

• mo_chip_premium now charges Missouri's MO HealthNet for Kids premium only when a CHIP child (is_chip_eligible_child) is in the tax unit. Missouri's pregnancy CHIP has no premium: CMS's approval of MO-15-0009 (pages 2 and 70) says targeted low-income pregnant women and unborn children "are not subject to premiums or cost-sharing charges", state plan section 8.2.4 says Show-Me Healthy Babies "does not have any cost-sharing requirements", and RSMo 208.640.1 sets the premium for parents and guardians of uninsured children. Without this change, postpartum mothers would newly trigger the premium. It also fixes main, which already charged it when a pregnant member was the only CHIP-eligible person.

is_chip_fcep_eligible_person keeps is_pregnant. The labels of is_chip_eligible_standard_pregnant_person, is_chip_eligible_pregnant and the PREGNANT_STANDARD CHIP category now say "pregnant or postpartum"; the enum member names and positions are unchanged.

Example (2025): a Virginia person between 60 days and 12 months postpartum, with income at 180% of the federal poverty line. That is above Virginia's 148% Medicaid pregnancy limit and below its 205% CHIP limit. On main, is_chip_eligible_standard_pregnant_person is false; with this PR it is true.

State dates

Each state's CHIP 12-month start date matches the Medicaid date that #10241 sets, so reusing the Medicaid parameter gives the right CHIP dates. Page numbers are PDF file pages.

State	CHIP pregnancy pathway	CHIP 12-month start	Source
CO	§2112 (CO-13-0015 CS8)	2022-07-01	CHIP SPA CO-22-0039
KY	§2112 (from 2022-07-01)	2022-04-01	CHIP SPA KY-22-0001
MO	§2112 (ME 95) plus unborn child (ME 96)	2023-07-07	CHIP SPA MO-23-0021
NJ	§2112 (NJ-13-0018 CS8)	2021-10-28	Section 1115 approval, Title XXI expenditure authority 1
RI	§2112 (CS8)	2022-10-01	CHIP SPA RI-22-0025
VA	FAMIS MOMS (section 1115)	2022-07-01	Section 1115 approval of 2021-11-18 (page 5: the postpartum expenditure authority, limited to April 1, 2022 through March 31, 2027 for lawfully residing enrollees; page 10: the operational ramp-up period); DMAS monitoring report, July 2021 to June 2022 (implemented July 1, 2022)
WV	§2112 (WV-19-0006)	2022-04-01	CHIP SPA WV-22-0017

Missouri's unborn-child mothers (ME 96) get only limited postpartum care: Show-Me Healthy Babies ends the day after hospital discharge, plus one postpartum visit (MO HealthNet manual 1855.030.15). The model does not represent that care, and FCEP keeps is_pregnant.

Dependencies and related work

• Date 12-month Medicaid postpartum coverage from each state's approval #10241 (merged as 0e659c8, now merged into this branch): dates the Medicaid postpartum parameter that CHIP now reads. The new tests use Colorado in 2021 (60 days), a Colorado person under 60 days postpartum in 2025, and Virginia in 2025 (365 days from 2022-07-01).
• Other health coverage: Apply the CHIP no-other-coverage test to pregnancy pathways in every state #10371 adds a check for other health coverage to is_chip_eligible_standard_pregnant_person. This PR does not include that check; the two merge without conflict.
• Income limits: Correct CHIP and Medicaid pregnancy income limits and effective dates #10365 corrects several dates and values in the CHIP pregnancy and FCEP income limits (KY, WV, CO, NJ, RI). Case 8's and Case 9's comments here hold under both versions. Both PRs append cases after Case 8 in is_chip_eligible_standard_pregnant_person.yaml, so whichever merges second keeps both blocks and renumbers its new cases.
• Correct Rhode Island pregnancy Medicaid routing and postpartum coverage #10391 (open, stacked on Correct CHIP and Medicaid pregnancy income limits and effective dates #10365) edits the same documentation line and appends cases to the same test file. It also adds ri_medicaid_postpartum_eligible to is_pregnant_for_medicaid_nfc, which both CHIP variables read after this PR. The PR that merges second resolves the conflict and re-checks Rhode Island postpartum routing in CHIP.
• Charge Texas, Michigan and Louisiana CHIP premiums only for CHIP children, not pregnancy-only (FCEP) members #10509 (follow-up): the Texas, Michigan and Louisiana CHIP premiums still trigger on any CHIP-eligible member, including pregnancy-only (FCEP) members.

Limitations

These match the model's Medicaid postpartum treatment:

• The model applies the income test throughout the postpartum period. In law, coverage continues regardless of income changes.
• Microsimulation datasets carry no postpartum inputs, so the postpartum extension has no effect in population runs (the same as Medicaid postpartum coverage). The Missouri premium change does take effect there.
• The day count approximates the end-of-month rule (Medicaid postpartum coverage should run to the end of the month in which the 60-day or 12-month period ends #10463).
• YEAR variables read January 1 values, so a mid-year start takes effect the next January.
• Virginia's STCs limit the 12-month extension for lawfully residing enrollees to March 31, 2027 (STC 17.c.iv). The Consolidated Appropriations Act, 2023 later made the federal option permanent, so whether the limit still applies is unresolved. The model has no end date (End Virginia's 12-month postpartum period for lawfully residing pregnant people on March 31, 2027 (FAMIS MOMS STC 17.c.iv) #10469).
• The model does not check that the person was enrolled while pregnant (SHO 21-007 page 3).

Test plan

• YAML tests whose expected values come from the statute and the state Medicaid and CHIP income limits. In is_chip_eligible_standard_pregnant_person.yaml:

  • Case 9: Colorado in 2021, covered at 60 days only.
  • Case 10: Virginia in 2025, covered for 12 months.
  • Case 11: a postpartum Colorado person keeps standard CHIP but not FCEP.
  • Case 12: Virginia in 2025, Medicaid (category PREGNANT) below its pregnancy limit and CHIP above it.
  • Case 13: Missouri in 2023, still 60 days (its July 7, 2023 date first applies in 2024).

  In is_chip_immigration_status_eligible.yaml, Case 15 covers a lawfully residing permanent resident 12 months postpartum in Virginia (2025), and Case 16 one in Colorado in 2022, where only the 60-day period applies. Two new cases in mo_chip_premium.yaml cover a tax unit whose only CHIP-eligible member is in the pregnancy pathway (premium $0): one with eligibility as an input, and a 2024 postpartum mother whose standard pregnancy eligibility is computed. I did not run tests locally; CI runs them.

• Source-conflict register entry: not applicable.

• ruff format --check passes on the changed Python files.

• Changelog fragment in changelog.d/

axiom: TheAxiomFoundation/rulespec-us#1932 queued

🤖 Generated with Claude Code

[PavelMakarchuk]

Program Review

PR #10363, "Extend standard pregnancy CHIP through the postpartum period" (hua7450). Round 1, full review.

IDs: C = critical, SA = should address, G = suggestion.

Paths: SP = policyengine_us/variables/gov/hhs/chip/is_chip_eligible_standard_pregnant_person.py, IM = policyengine_us/variables/gov/hhs/chip/is_chip_immigration_status_eligible.py, MP = policyengine_us/variables/gov/states/mo/hhs/chip/mo_chip_premium.py, TS, TI and TM = the matching test files for SP, IM and MP under policyengine_us/tests/policy/baseline/gov/hhs/chip/ and .../gov/states/mo/hhs/chip/, P = policyengine_us/parameters/gov/hhs/medicaid/eligibility/categories/pregnant/postpartum_coverage.yaml. Line numbers refer to the reviewed head. Every PDF link uses the file page number.

Source Documents

• Reviewed head SHA: 0e7f410
• Mode: full
• Scope: merge base 0e659c8 (merged #10241) to the head. 7 files, 195 insertions and 8 deletions: 3 variables, 3 test files and a changelog fragment. No parameter changes.
• Statute: 42 USC 1397ll(d)(2)(A), 42 USC 1397gg(e)(1)(K) with its amendment notes, and 42 CFR 457.342 (also 457.310 and 457.340, which have no postpartum text).
• PDF: CMS SHO #21-007 (pages 2-7, 17 and 18; #page=7, #page=17).
• PDF: CHIP approvals CO-22-0039 (page 1), KY-22-0001, MO-23-0021, RI-22-0025 and WV-22-0017 (page 2 each).
• PDF: Virginia FAMIS MOMS 1115 amendment of 2021-11-18 (pages 1, 5), DMAS monitoring report, July 2021 to June 2022 (page 11), and the New Jersey 1115 approval of 2021-10-28 (pages 1, 9; Wayback capture, see G7).
• PDF: Missouri MO-15-0009 approval and state plan (page 2 read as an image, page 70).
• Web: RSMo 208.640 and MO HealthNet manual 1855.030.15.
• Repository: is_pregnant_for_medicaid_nfc.py:10-16, P (CO, KY, MO, NJ, RI, VA, WV), parameters/gov/hhs/chip/pregnant/income_limit.yaml:24-26 and parameters/gov/hhs/chip/fcep/income_limit.yaml.
• Year: the new tests cover 2021 and 2025.

Branch Status

• The head is a merge of main at 0e659c8, which is merged #10241. The branch is 5 commits ahead and 14 behind main, and it merges cleanly. No later commit on main touches CHIP, the Medicaid pregnancy category, the postpartum inputs or the Missouri CHIP files.
• This PR is the follow-up to #10241: CHIP now reads the per-state dates that #10241 set in P. It fixes issue #10362 (open). Issues #10463 (month-end rule) and #10469 (Virginia end date) are open and are named in the Limitations.
• Overlaps: #10371 merges with no conflict, and #10365 conflicts only where both PRs append to TS, as the body says. #10391 conflicts and is not in the body (G3).

What the PR gets right

• Legal design. 42 USC 1397gg(e)(1)(K) makes separate CHIP follow the state's Medicaid 12-month election by law, "not at the option of the State". Reading is_pregnant_for_medicaid_nfc is therefore correct and avoids a duplicate CHIP parameter. For all 7 states with a modeled standard pregnancy pathway, the CHIP SPA or 1115 effective date equals the Medicaid date in P.
• Citation. SP:16 uses (K), the current letter after Pub. L. 119-21 redesignated the ARPA text. SHO #21-007 and the CMS letters still use the old letter (J).
• FCEP. is_chip_fcep_eligible_person keeps is_pregnant. SHO #21-007 pages 4 and 7 exclude unborn-child coverage from the extension.
• Missouri premium. MP:38 matches RSMo 208.640.1 and the MO-15-0009 exemption (page 2 and section 8.2.4 on page 70). It also fixes main, which charged the premium when a pregnant member was the only CHIP-eligible person.
• Tests. Reverting any of the three changes fails a new case (Cases 9-12, Case 15 and the new Missouri case). For the standard pathway, Case 9 fails if 12 months applies everywhere, and Cases 10 and 12 fail if 60 days applies everywhere.
• Code. No hard-coded values, YEAR-to-YEAR reads, no cycle; formatting and changelog correct.

Critical (Must Fix)

None.

Should Address

SA1. The lawfully residing pregnancy branch has no 60-day case.

• Where: IM:54-60; the only postpartum test is Case 15 (TI:303-330).
• Evidence: Case 15 tests Virginia in 2025 (365 days). Two wrong versions of the branch pass every related test: "pregnant or under 60 days or under 12 months postpartum", and "pregnant or under 12 months postpartum". The related tests are the CHIP folder plus every test file that sets a postpartum input or reads this variable (Medicaid pregnancy and lawfully residing, CA CCHIP, MS HMW, WIC): 189 tests. SHO #21-007 Q11 (#page=17) gives 12 months to these enrollees only in a state that "has also elected" the extension. Both wrong versions drop that condition. They would also make the branch ignore a reform of P.
• Why this severity: a missing edge case for an already-tested variable. The formula is correct.
• Fix: append to TI. It passes at the head and fails under both wrong versions and under a Colorado date moved to 2022-01-01. A Virginia version also works, but it does not catch the Colorado date.

# The 12-month extension reaches lawfully residing pregnant people only in a
# state that has also elected it (SHO 21-007 Q11). Colorado's 12-month
# extension took effect July 1, 2022 (CHIP SPA CO-22-0039); the annual model
# reads January 1, so 2022 still uses the 60-day period.
- name: Case 16, Colorado pregnancy option covers a recent permanent resident 60 days postpartum, not 12 months, in 2022.
  period: 2022
  input:
    people:
      person1:
        age: 30
        is_pregnant: false
        under_60_days_postpartum: true
        immigration_status: LEGAL_PERMANENT_RESIDENT
        years_since_us_entry: 2
      person2:
        age: 30
        is_pregnant: false
        under_12_months_postpartum: true
        immigration_status: LEGAL_PERMANENT_RESIDENT
        years_since_us_entry: 2
    households:
      household1:
        members: [person1]
        state_code: CO
      household2:
        members: [person2]
        state_code: CO
  output:
    is_chip_immigration_status_eligible: [true, false]

Suggestions

G1. Pin the first year of the Colorado and Missouri extensions.

• Where: Cases 9-12 (TS:192-304) use only 2021 and 2025. Dates: P:154-156 (CO) and P:211-213 (MO).
• Evidence: If Colorado's 2022-07-01 date moves to 2022-01-01, or Missouri's 2023-07-07 date moves to 2024-07-07, all 189 related tests still pass, the Medicaid tests included. KY, RI and WV are in the same position. The January 1 read has no CHIP test.
• Why a suggestion: the dates belong to the merged #10241 parameter, and the Medicaid tests have the same gap. The CHIP read of the date is pinned: a later Virginia date fails Cases 10, 12 and 15. Both dates match the SPA text.
• Fix: these cases pass at the head. Case 13 fails on the Colorado date change. The Missouri 2024 case fails on the Missouri date change. It also fails if MP goes back to is_chip_eligible (premium 1,548 instead of 0), and it computes eligibility, while the new Missouri case sets it as an input. If you add SA1, Colorado is already pinned, and the Missouri 2024 case is enough. Append to TS:

# Colorado's 12-month extension took effect July 1, 2022 (CHIP SPA CO-22-0039).
# The annual model reads January 1, so the extension starts in 2023.
- name: Case 13, Colorado standard pregnancy CHIP does not yet cover 12 months postpartum in 2022.
  period: 2022
  input:
    people:
      person1:
        age: 30
        is_pregnant: false
        under_12_months_postpartum: true
        medicaid_income_level: 2.3
    households:
      household:
        members: [person1]
        state_code: CO
  output:
    is_chip_eligible_standard_pregnant_person: false

- name: Case 14, Colorado standard pregnancy CHIP covers 12 months postpartum in 2023.
  period: 2023
  input:
    people:
      person1:
        age: 30
        is_pregnant: false
        under_12_months_postpartum: true
        medicaid_income_level: 2.3
    households:
      household:
        members: [person1]
        state_code: CO
  output:
    is_chip_eligible_standard_pregnant_person: true

Append to TM (the 2023 case can go in TS instead):

# Missouri's 12-month extension took effect July 7, 2023 (CHIP SPA MO-23-0021).
# The annual model reads January 1, so the extension starts in 2024.
- name: Missouri standard pregnancy CHIP does not yet cover 12 months postpartum in 2023.
  period: 2023
  input:
    people:
      person1:
        age: 30
        is_pregnant: false
        under_12_months_postpartum: true
        medicaid_income_level: 2.3
    households:
      household:
        members: [person1]
        state_code: MO
  output:
    is_chip_eligible_standard_pregnant_person: false

# Show-Me Healthy Babies members pay no premium (MO-15-0009), so a postpartum
# mother whose CHIP eligibility is computed through the pregnancy pathway does
# not trigger the MO HealthNet for Kids premium.
- name: Missouri postpartum mother in standard pregnancy CHIP pays no premium in 2024.
  period: 2024
  input:
    people:
      person1:
        age: 30
        is_pregnant: false
        under_12_months_postpartum: true
        medicaid_income_level: 2.3
    tax_units:
      tax_unit:
        members: [person1]
        tax_unit_medicaid_income_level: 2.3
    households:
      household:
        members: [person1]
        state_code: MO
  output:
    is_chip_eligible_standard_pregnant_person: true
    chip_category: PREGNANT_STANDARD
    mo_chip_premium: 0

G2. Two test comments state #10365's values, not this branch's values.

• Where: TS:187-188 and TS:192-196.
• Evidence: Line 188 now says Colorado has offered FCEP "since January 1, 2025". On this branch, parameters/gov/hhs/chip/fcep/income_limit.yaml gives Colorado 2023-10-30: 2.65, which the replaced text matched. Case 9 says the CHIP limit is 260% "plus the 5 percentage point MAGI disregard", that is 265%. On this branch the 2021 limit is 2.60 (parameters/gov/hhs/chip/pregnant/income_limit.yaml:24-26), and medicaid_income_eligible adds no disregard, so the model applies 260%. Both comments match #10365, as the body says.
• Why a suggestion: comments only. Income of 230% passes under both values.
• Fix: merge #10365 first and say so in the body. Or use wording that holds on either branch: keep "(Colorado since October 30, 2023)" at line 188 until #10365 merges, and end the Case 9 comment with "Income of 230% FPL is above Colorado's Medicaid pregnancy limit and below its CHIP pregnancy limit."

G3. The body does not name #10391, which conflicts with this PR and changes the variable that CHIP now reads.

• Where: SP:11 and SP:58-64; TS; the "Dependencies and related work" section of the body.
• Evidence: a three-way merge with #10391's head gives 1 conflict in SP (the documentation line; the return block merges cleanly) and 7 conflict hunks in TS. #10391 also adds ri_medicaid_postpartum_eligible to is_pregnant_for_medicaid_nfc. After both merge, Rhode Island's Medicaid continuation flows into both CHIP variables. No wrong outcome was found, because CHIP still requires ~is_medicaid_eligible. #10391 itself was not reviewed.
• Fix (body): "#10391 (open, stacked on #10365) edits the same documentation line and appends cases to the same test file. It also adds ri_medicaid_postpartum_eligible to is_pregnant_for_medicaid_nfc, which both CHIP variables read after this PR. The PR that merges second resolves the conflict and re-checks Rhode Island postpartum routing in CHIP."

G4. Two downstream labels still say "pregnant" only.

• Where: policyengine_us/variables/gov/hhs/chip/is_chip_eligible_pregnant.py:7-8 and policyengine_us/variables/gov/hhs/chip/chip_category.py:6.
• Evidence: both now cover a person up to 12 months postpartum (Case 11 asserts PREGNANT_STANDARD for one). No code or test reads these strings.
• Fix: keep the enum member name and position (chip_category.py:9-10 relies on stable indices).

    label = "Pregnant or postpartum person eligible for CHIP"
    documentation = "Determines if a pregnant or postpartum person is eligible for the Children's Health Insurance Program through either the standard pregnant pathway or FCEP"

    PREGNANT_STANDARD = "Pregnant or postpartum - Standard"

G5. The name has_chip_member no longer matches its value.

• Where: MP:38 and MP:92.
• Evidence: the variable now means "the tax unit has a CHIP child", which is the RSMo 208.640.1 trigger.
• Fix: has_chip_child = add(tax_unit, year, ["is_chip_eligible_child"]) > 0 at line 38, and return has_chip_child * monthly_premium at line 92.

G6. Say in the body that the extension has no effect in microsimulation.

• Where: the Limitations section of the body.
• Evidence: the default dataset (populace_us_2024.h5) has is_pregnant but no postpartum input, so under_60_days_postpartum and under_12_months_postpartum are always false there. Population runs therefore show no postpartum CHIP enrollment or cost in CO, KY, MO, NJ, RI, VA and WV. The PR changes no default, and Medicaid postpartum coverage has the same gap. The Missouri premium change does take effect: it lowers chip_premium for Missouri tax units whose only CHIP member is pregnant (a small effect).
• Fix (body): "Microsimulation datasets carry no postpartum inputs, so the postpartum extension has no effect in population runs (the same as Medicaid postpartum coverage)."

G7. The New Jersey source link in the body is dead.

• Where: the NJ row of the "State dates" table.
• Evidence: the nj.gov URL returns "404 - Page Not Found". A Wayback capture of 2025-10-31 (483 pages) shows that page 9 is the title XXI expenditure authority, so the claim is correct.
• Fix: link an archived copy, or the medicaid.gov copy that P:83 already cites (.../1115/downloads/nj/nj-1115-request-ca.pdf, also 483 pages). Check page 9 on that copy before you change the anchor. No code reference changes.

Investigated and cleared

• CHIP dates. For CO, KY, MO, NJ, RI, VA and WV, the CHIP SPA or 1115 effective date equals the date in P.
• 60-day years. The model keeps the CHIP income test in 60-day years. That matches the law: without the 12-month election, CHIP has no continuous-eligibility rule for pregnant enrollees (SHO #21-007 page 18, Q14).
• Period and cycles. Both callers are YEAR and read the YEAR NFC with period. No Medicaid variable reads a CHIP variable, so there is no cycle.
• Shared immigration flag. is_chip_immigration_status_eligible also feeds is_chip_eligible_child, ca_cchip_eligible and ms_hmw_other_coverage_ineligible. No current state lets a postpartum lawfully residing minor into the child pathway. A Rhode Island 17-year-old permanent resident 12 months postpartum passes through the pregnancy branch, which SHO Q11 supports.
• Other CHIP premiums. The LA, FL, DE, KS, TX and MI premiums still read is_chip_eligible, but none of those states has a standard pregnancy pathway, so the extension does not reach them.
• Existing Missouri cases. They already set is_chip_eligible_child, so their results do not change.
• axiom: line. TheAxiomFoundation/rulespec-us#1932 queued points to an open issue titled for this PR. Process note only; not counted.

Out of scope / follow-up

• TX, MI and LA premiums. These states have FCEP limits above 0 and their premiums read is_chip_eligible, so a pregnancy-only FCEP member can trigger a children's premium. This is the pattern this PR fixes for Missouri. It exists on main, and the state premium law was not checked. Consider a follow-up issue.
• Kentucky start. parameters/gov/hhs/chip/pregnant/income_limit.yaml:49-52 starts Kentucky's pregnancy limit in 2021, but the section 2112 pathway starts 2022-07-01. #10365 corrects it. Until then, this PR also extends that early coverage to postpartum persons.
• Lawfully residing start. The NJ and VA 1115s start the extension for lawfully residing enrollees on 2022-04-01. Under the January 1 read, NJ gives that group 12 months in annual 2022. Merged #10241 does the same, and the population is very small. A limitation line is optional.
• Continuous eligibility. In law, a CHIP enrollee stays in CHIP for all 12 months even if income falls into the Medicaid range (SHO pages 5-6, Q15). The model moves that person to Medicaid; coverage is the same. The body's Limitations cover this.

Validation Summary

Check	Result
CI	33/33 pass at 0e7f4104c7
Local tests	3 changed files 45/45 (12, 15, 18); gov/hhs/chip 150 pass; related CHIP, Medicaid, CCHIP, HMW and WIC tests 189 pass
Partner contract tests	None edited. Read-only run of partners/analytics_coverage/edge_cases/federal/healthcare: 70/70
References	6 new code references, all corroborate (SHO pages 2, 4, 7, 17; MO-15-0009 pages 2 and 70). 9 body sources: 8 corroborate, NJ link dead (G7)
Regulatory	Design correct; 7 of 7 state dates match; Missouri premium matches state law and the state plan
Code patterns	No hard-coded values, YEAR-to-YEAR, add() aggregation, formatted, changelog correct
Mutation	17 changes to the new logic, the shared NFC test and the date parameter: 14 caught, 3 not caught (SA1, G1). The proposed cases catch all 3
Merge trials	#10371 clean; #10365 test file only (expected); #10391 variable and test file (G3)
Microsimulation	No postpartum inputs in the default dataset; the extension has no population effect (G6)
Branch	5 ahead / 14 behind main; merges cleanly
Other reviews	None

Review Severity: REQUEST_CHANGES

One should-address item is open: SA1, a missing 60-day test for the lawfully residing branch. Seven suggestions are open (G1 to G7).

Next Steps

To auto-fix issues: run the fix-pr workflow for this PR.

1. Append Case 16 from SA1 to TI.
2. Take the suggestions that you agree with. G1 is pre-run test YAML. G2 changes comments only. G3, G6 and G7 change the body only. G4 and G5 are label and name changes.
3. Run make format, then run the three changed test files and policyengine_us/tests/policy/baseline/gov/hhs/chip.
4. Agree a merge order with #10365 and #10391. The PR that merges second keeps both test blocks in TS and renumbers its cases.

[hua7450]

Thanks, Pavel. All items are applied at 3079297 (fix commit 3669616, then a merge of main; make format changed nothing). Expected values were checked by hand; CI runs the tests.

Item	Disposition
SA1	Done. Appended your Case 16 to is_chip_immigration_status_eligible.yaml: Colorado in 2022, a recent permanent resident under 60 days postpartum is eligible and one past 60 days is not. Colorado's 2022-07-01 date first applies in 2023.
G1	Done, without the separate Colorado cases, since Case 16 already pins Colorado's 2022 year. Your Missouri 2023 case is Case 13 in is_chip_eligible_standard_pregnant_person.yaml, the file for the variable it asserts. The Missouri 2024 premium case is appended to mo_chip_premium.yaml. It computes standard pregnancy eligibility and asserts PREGNANT_STANDARD with premium 0.
G2	Done. Case 8 keeps "(Colorado since October 30, 2023)", so #10365's change to that line merges cleanly. Case 9 now ends "Income of 230% FPL is above Colorado's Medicaid pregnancy limit and below its CHIP pregnancy limit."
G3	Done. The body's Dependencies section now names #10391 with your wording.
G4	Done. is_chip_eligible_pregnant label and documentation, and the PREGNANT_STANDARD label ("Pregnant or postpartum - Standard"). Member names and positions are unchanged; the partner CHIP test that reads the index (1) is unaffected.
G5	Done. has_chip_child in mo_chip_premium.
G6	Done. Limitations now say the extension has no effect in population runs.
G7	Done. The NJ row now links the medicaid.gov copy (nj-1115-request-ca.pdf, 483 pages). I checked that its page 9 is "Title XXI EXPENDITURE AUTHORITY 1. Expenditures for Benefits for Postpartum Women", starting April 1, 2022 for CHIP lawfully residing pregnant women.
Out of scope: TX, MI, LA premiums	Filed #10509.
Merge order	#10365 is planned to merge first. This PR will then keep both test blocks in is_chip_eligible_standard_pregnant_person.yaml and renumber. #10371 merges cleanly. #10391 conflicts in the same test file and on the variable's documentation line.

The Axiom parity issue (TheAxiomFoundation/rulespec-us#1932) is updated to this head and the new cases.