Skip to content

Count New Jersey foster caretakers and parents of 18-year-old students in the Medicaid parent group - #10984

Merged
hua7450 merged 4 commits into
mainfrom
nj-medicaid-caretaker-definition
Nov 17, 2026
Merged

hua7450 merged 4 commits into
mainfrom
nj-medicaid-caretaker-definition

Conversation

@hua7450

@hua7450 hua7450 commented Nov 15, 2026 •

Copy link
Copy Markdown
Collaborator

Summary

Part of #10775: this PR covers New Jersey, and the issue stays open for other states.

New Jersey's approved state plan page S25 (NJ-13-0011-MM1, approved 02/28/2014, effective 01/01/2014) checks two options for the parent and caretaker relative group (42 CFR 435.110) that the model did not encode:

  • "The definition of caretaker relative includes any adult with whom the child is living and who assumes primary responsibility for the dependent child's care" (the 42 CFR 435.4 paragraph (3) option).
  • "This eligibility group includes individuals who are parents or other caretakers of children who are 18 years old, provided the children are full-time students in a secondary school or the equivalent level of vocational or technical training."

is_parent_for_medicaid_nfc dropped every child with is_in_foster_care from the dependent-child test, so New Jersey adults caring for foster children, including grandmothers caring for foster grandchildren, got the adult group. It also left New Jersey's student option off.

Legal basis

  • 42 CFR 435.4, "Caretaker relative", paragraph (3): "At State option, ... an adult with whom the child is living and who assumes primary responsibility for the dependent child's care."
  • New Jersey approved state plan S25, NJ-13-0011-MM1: both boxes checked.
  • N.J.A.C. 10:69-3.13(a) also covers a child "under age 19 and a full-time student in a secondary school, or in the equivalent level of vocational or technical training".
  • No later New Jersey amendment replaces S25. Medicaid.gov tags 19 New Jersey amendments approved after NJ-13-0011-MM1 with "Eligibility". Seven have summaries on other topics, and none of the other 12 contains a new S25 page. NJ-23-0005's coverage summary lists the parent and caretaker relative group as "CONVERTED" and not included in that package.

Changes

File Change
parameters/.../categories/parent/caretaker_relative/any_adult_option.yaml (new) Per-state boolean for the 42 CFR 435.4 paragraph (3) option. New Jersey true from 2014-01-01, citing S25. The other 50 states stay false, which means not yet researched, not a verified non-election.
variables/.../categories/parent/is_parent_for_medicaid_nfc.py In states with the option, a foster child counts as a dependent child. Elsewhere the formula keeps excluding foster children as a proxy for an unrelated foster placement. Comments updated.
parameters/.../categories/parent/dependent_child/student_option.yaml New Jersey true from 2014-01-01 (main had false, an unverified default), citing S25 and N.J.A.C. 10:69-3.13(a).
parameters/.../categories/parent/requires_deprivation.yaml New Jersey false from 2014-01-01 (was 2018-01-01), citing S25, which eliminates the deprivation requirement.
source_conflicts/candidates.yaml Records us-nj:medicaid-caretaker-relative-any-adult (below), with its ruling record.
tests/.../categories/parent/is_parent_for_medicaid_nfc.yaml (new) Four New Jersey cases.

Results

is_parent_for_medicaid_nfc for the adult in each test household, 2026:

Case Household Main This PR
1 NJ adult, $3,600 of earnings, caring for an 8-year-old in foster care (the partner's report in the issue) false true
2 NJ grandmother, $16,557 of earnings, caring for her 8-year-old grandchild in foster care false true
3 NJ parent, $33,114 of earnings, with an 18-year-old full-time high school student false true
4 NJ parent, $33,114 of earnings, with an 18-year-old who is not a student false false
— Missouri foster parent and Missouri foster grandmother (existing cases in is_parent_for_medicaid.yaml, which assert is_parent_for_medicaid) false false

In Case 1, $3,600 is 17% of the 2026 poverty guideline for two people ($21,640), under the model's 34% parent group limit for New Jersey, so medicaid_category moves from ADULT to PARENT; the case asserts [PARENT, OLDER_CHILD]. A New Jersey foster caretaker with income above the parent group's limit stays in the adult group, so Medicaid eligibility for adults aged 19 to 64 doesn't change. New Jersey doesn't cover parents through the medically needy pathway, so that pathway doesn't change.

Source conflict

N.J.A.C. 10:69 (AFDC-related Medicaid, readopted effective July 15, 2024) confines the parent role to listed relatives. 10:69-1.5 defines a dependent child as one living "with a parent or other enumerated relative", and 10:69-3.10(a)1 lists the relatives (grandparents, siblings, step-relatives, aunts, uncles, first cousins, nephews and nieces). S25 elects the any-adult option.

The project lead ruled to follow S25. A state elects the 42 CFR 435.4 paragraph (3) option through its state plan, and S25 is the federally approved description of New Jersey's parent and caretaker relative group. N.J.A.C. 10:69 also keeps the deprivation test (10:69-3.15), which S25 removes and the model already leaves out for New Jersey, so the chapter still carries rules the plan has replaced. Grandmothers and other relatives caring for a foster child count as caretakers under both texts.

The ruling record is on this PR. This PR records the conflict as the ruled candidate us-nj:medicaid-caretaker-relative-any-adult in source_conflicts/candidates.yaml, with no marker yet (CONTRIBUTING.md, "Source conflicts", step 1).

Not changed

Tests

  • 4 new YAML cases in is_parent_for_medicaid_nfc.yaml; existing cases unchanged. Expected values traced by hand from 42 CFR 435.4 and S25; not run locally.
  • CI passes

Review fixes

From DTrim99's review, in commit 29cbb14:

  • The candidate now carries ruling: {by: PolicyEngine, date: '2026-11-15', record: …} with the project lead's ruling comment, lists this PR under pe_us.prs, and cites N.J.A.C. 10:78-1.1(a) (NJ FamilyCare covers "parents, caretakers and caretaker relatives").
  • requires_deprivation: New Jersey false from 2014-01-01, citing S25 page 31.
  • Case 1 asserts medicaid_category: [PARENT, OLDER_CHILD]; Case 4 adds a New Jersey 18-year-old who is not a student (parent false).
  • rulespec-us 2261 now uses Case 4 as companion row 4, says deprivation is eliminated from 2014 to match, and cites head 29cbb14.
  • Other states' S25 pages stay deferred to Medicaid parent group: count foster and kinship caretakers who meet the 42 CFR 435.4 caretaker definition #10775.

axiom: TheAxiomFoundation/rulespec-us#2261 queued

🤖 Generated with Claude Code

…s in the Medicaid parent group

New Jersey's approved state plan S25 (NJ-13-0011-MM1) elects the 42 CFR
435.4 paragraph (3) option counting any adult who lives with a dependent
child and assumes primary responsibility for the child's care as a
caretaker relative, and the 18-year-old full-time student option.

- Add a per-state any-adult caretaker option parameter, true for New
  Jersey from 2014; where elected, foster children count as dependent
  children in is_parent_for_medicaid_nfc.
- Set New Jersey's dependent-child student option to true from 2014.
- Record the S25 / N.J.A.C. 10:69 conflict as a ruled candidate.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@hua7450
hua7450 requested a review from DTrim99 November 16, 2026 18:46
@DTrim99

DTrim99 commented Nov 16, 2026

Copy link
Copy Markdown
Collaborator

PR Review: Count New Jersey foster caretakers and parents of 18-year-old students in the Medicaid parent group

Summary

The policy change is correct. I checked the federal rule first, then New Jersey's own texts.

  • Federal rule. 42 CFR 435.4 (eCFR, current) makes both elections state options:
    • "Caretaker relative", paragraph (3): "At State option, ... or an adult with whom the child is living and who assumes primary responsibility for the dependent child's care."
    • "Dependent child", paragraph (1): "Is under the age of 18, or, at State option, is age 18 and a full-time student in secondary school (or equivalent vocational or technical training), if before attaining age 19 the child may reasonably be expected to complete such school or training."
  • New Jersey elected both. I rendered page 31 of NJ-13-0011-MM1 (TN 13-0011-MM1, approval date 02/28/2014, effective date 01/01/2014). Both boxes are checked:
    • the 18-year-old full-time student option;
    • "The definition of caretaker relative includes any adult with whom the child is living and who assumes primary responsibility for the dependent child's care".
    • The domestic-partner and other-relative boxes are unchecked, and the deprivation requirement is eliminated.
  • The N.J.A.C. 10:69 quotes are on the cited PDF pages. That chapter was readopted effective July 15, 2024.
    • 10:69-3.13(a) agrees on the student option.
    • 10:69-1.5 and 3.10(a)1 confine the parent role to listed relatives, which is the conflict the PR records.
  • Backdating and keys. Both NJ values start on 2014-01-01, matching the S25 effective date. The new parameter has the same 51-key set as student_option, age_limit and income_limit, which the formula already indexes the same way.

One item needs fixing: the new candidates.yaml entry is marked ruled but doesn't link the project lead's ruling record.

Critical (must fix)

None.

Should address

  1. source_conflicts/candidates.yaml, us-nj:medicaid-caretaker-relative-any-adult: state: ruled with no linked ruling record.

Suggestions

  1. candidates.yaml: pe_us.prs: []. List Count New Jersey foster caretakers and parents of 18-year-old students in the Medicaid parent group #10984, the ruling PR, so step 2 can pin the merge commit. Three other ruled entries leave prs empty, but most list their PR.
  2. Add a New Jersey negative student case. An 18-year-old who is not a student should give a parent result of false. rulespec-us#2261 companion row 4 says "no PE-US New Jersey case" and points to the Missouri case. A New Jersey row would let the companion tests come entirely from this file.
  3. Pin the reported symptom. The partner's report was about medicaid_category (ADULT instead of PARENT). A medicaid_category assertion on Case 1 would cover that end to end. By hand trace it gives PARENT, since $3,600 / $21,640 = 16.6%, under NJ's 0.34 limit plus the 0.05 disregard.
  4. Other states are cheap to source. Several states' S25 pages are already cited in student_option.yaml, and the any-adult box sits on the same page:
  5. requires_deprivation NJ starts 2018-01-01. The same S25 page eliminates deprivation from 2014-01-01, and rulespec-us#2261 required output 3 says "true for New Jersey from January 1, 2014". Either backdate NJ to 2014-01-01, citing S25 page 31, or scope the issue's output to 2018 onward. The impact is low because NJ income_limit also starts in 2018.

Verified correct

  • Federal definitions. The 42 CFR 435.4 "Caretaker relative" (1) to (3) and "Dependent child" (1) to (2) text matches the PR body, the parameter references, the candidate's readings.statute and the rulespec issue.
  • S25 page 31. It is PDF page 31 of NJ-13-0011-MM1, so the #page=31 anchor is correct. The page is a scan with no text layer, so I checked it visually.
    • The student option and the any-adult option are checked.
    • The domestic-partner and other-relatives options are unchecked.
    • The option eliminating deprivation is selected.
    • The page header cites 42 CFR 435.110, 1902(a)(10)(A)(i)(I) and 1931(b) and (d).
  • N.J.A.C. 10:69 quotes (PDF pages in the nj.gov compilation):
    • Chapter notes, readopted "Effective: July 15, 2024" (page 1).
    • "Caretaker relative" means "the legally responsible adult or adults residing with the children" (page 9).
    • "Dependent child" means a child "living in New Jersey with a parent or other enumerated relative" (page 10).
    • "Parent-person" (page 12).
    • 10:69-3.10(a)1 lists the relatives (page 64).
    • 10:69-3.13(a): "under age 18, or under age 19 and a full-time student in a secondary school, or in the equivalent level of vocational or technical training", expected to finish before 19 (page 69).
    • 10:69-3.15, "Deprivation of parental support or care" (page 72).
  • Candidate citation format. The third source's '… #page 9-10,12,64' follows the CLAUDE.md multi-page title convention. It is quoted, and the href opens the first page.
  • Hand-checks (2026 unless noted):
    1. Case 1 foster parent = true. Age 8 < NJ age_limit 18. in_foster_care is true, but any_adult_option[NJ] is true, so excluded_foster_child is false. The child is a tax dependent, so is a dependent child. The adult is head and not a dependent. requires_deprivation[NJ] is false, so the result is true.
    2. Case 1 foster child = false: not head or spouse.
    3. Case 2 grandmother = true, by the same path. Income 20 × 52 × $15.92 = 1,040 × 15.92 = $16,556.80, and NJ's 2026 minimum wage of $15.92 is confirmed.
    4. Case 2 grandchild = false.
    5. Case 3 parent = true. 18 < 18 fails, but student_option[NJ] is true, the child is a full-time student, and 18 < student_age_limit 19. Income 2,080 × $15.92 = $33,113.60.
    6. Case 3 student = false: a dependent, not a filer.
    7. Missouri foster parent (existing case, 2025) = false. any_adult_option[MO] is false, so the foster child is excluded, there is no dependent child, and nfc is false. Income of 0.1 FPL is under Missouri's limit, so this case pins the ~elects_any_adult term: dropping the foster exclusion everywhere would flip it.
    8. Missouri foster grandmother (existing) = false, by the same path.
    9. PR-body arithmetic. The 2026 poverty guideline for 2 is $15,960 + $5,680 = $21,640, and $3,600 / $21,640 = 16.6% ("17%"). That is under NJ's 0.34 limit, so the category is PARENT by trace.
    10. Mutation checks. Flipping NJ any_adult_option to false fails Cases 1 and 2. Flipping NJ student_option back to false fails Case 3. Every new branch and value is pinned.
  • Formula.
    • It is vectorized: string-keyed parameter indexing plus .astype(bool), the same as student_option.
    • Entity and period are correct: a Person-level YEAR variable, with the state read from the household.
    • It reinvents no existing variable.
    • Callers are is_parent_for_medicaid and is_in_medicaid_medically_needy_category. NJ medically needy parent is false, so the medically needy pathway is unchanged, as the PR body says.
  • NH Add New Hampshire state supplementation for 2026 #10156 pattern. The key comes from state_code_str, not from a state-gated variable, and all 51 keys are present, so multi-state simulations look up valid keys.
  • Housekeeping.
    • The changelog fragment changelog.d/nj-medicaid-caretaker-definition.fixed.md is at top level with type fixed.
    • The tests are in the variable-named file is_parent_for_medicaid_nfc.yaml, with plain periods.
    • No partner-test files are touched.
  • Source-conflict guard. The candidate has every required field (id, program, question, years, three sources, readings, pe_us with follows, code and prs, and state). Both pe_us.code paths are tracked files, and no marker is added (step 1). The guard passes at the head and on the tree merged with main.

Axiom line assessment

axiom: TheAxiomFoundation/rulespec-us#2261 queued is valid.

  • Status. The issue is OPEN and labelled pe-parity. It was opened 2026-11-15 by the author.
  • Dispatch-ready.
    • It gives target module paths: us/regulations/42-cfr/435/4.yaml and the us-nj/.../nj-13-0011-mm1.yaml S25 module.
    • It gives corpus citations and quotes the law verbatim. I checked the quotes against eCFR and the S25 scan.
    • It lists the readings taken, including the plan-over-10:69 conflict, the required outputs, and companion tests derived from 435.4 and S25 rather than from PolicyEngine output.
    • It records the blocker: the pinned corpus release has no New Jersey Medicaid state plan scope.
  • Matches the head. The issue cites head 00af8aa94f. The current head 9d12aa73a3 only merges main on top of it, and the PR's files and candidate entry are byte-identical between the two.
  • One mismatch (Suggestion 5). Required output 3 says deprivation is eliminated from 2014, while the PE-US NJ value starts in 2018.

CI status

All 38 checks pass at head 9d12aa73a3 (run created 2026-11-16 03:09 UTC), including PR validation (lock, guard tests, lint, changelog), Household API Partners, all baseline, contrib and Python shards, and Microsimulation. Re-checked at the end of this review; nothing is pending.

Branch status

Verdict: REQUEST_CHANGES

@DTrim99 DTrim99 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: 0 critical, 1 should address (the ruled candidate us-nj:medicaid-caretaker-relative-any-adult needs a linked ruling record, or state: open), 5 suggestions. Full report in the comment above. The policy itself checks out against 42 CFR 435.4 and NJ S25.

@hua7450

hua7450 commented Nov 16, 2026

Copy link
Copy Markdown
Collaborator Author

Source-conflict ruling (project lead, 2026-11-15): us-nj:medicaid-caretaker-relative-any-adult

New Jersey's caretaker relative definition follows the approved state plan page S25 (NJ-13-0011-MM1, approved 2014-02-28, effective 2014-01-01). That page elects 42 CFR 435.4 option (3): "any adult with whom the child is living and who assumes primary responsibility for the dependent child's care". It takes precedence over N.J.A.C. 10:69-1.5 and 10:69-3.10(a)1, which list relatives only.

hua7450 and others added 2 commits November 16, 2026 15:11
- us-nj:medicaid-caretaker-relative-any-adult: add the ruling record
  (by PolicyEngine, 2026-11-15, the project lead's comment on this PR),
  list this PR under pe_us.prs, and cite N.J.A.C. 10:78-1.1(a), which
  covers "parents, caretakers and caretaker relatives".
- requires_deprivation: New Jersey false from 2014-01-01, citing S25
  (NJ-13-0011-MM1, PDF page 31), which eliminates the deprivation
  requirement from that date.
- is_parent_for_medicaid_nfc.yaml: assert medicaid_category
  [PARENT, OLDER_CHILD] on Case 1, the symptom the partner reported
  ($3,600 / $21,640 = 16.6%, under the 0.34 parent and 1.47 older-child
  limits), and add Case 4, a New Jersey 18-year-old who is not a student
  (parent false).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@hua7450
hua7450 requested a review from DTrim99 November 16, 2026 20:16
@DTrim99

DTrim99 commented Nov 17, 2026

Copy link
Copy Markdown
Collaborator

PR Review (round 2): Count New Jersey foster caretakers and parents of 18-year-old students in the Medicaid parent group

Reviewed at head 29cbb14acd. Round 1 (2026-11-16) was at 9d12aa73a3 and requested changes: 0 critical, 1 should, 5 suggestions. Since then the branch has one merge of main, 77041f09fc, and one fix commit, 29cbb14acd. The merge had no hand-resolved conflicts: its remerge diff is empty, and none of the PR's own files changed in it. I ran no tests or Python this round. The results below come from reading the head, hand traces, the sources and the CI run at this head.

Summary

The fix commit closes the one open item. It also takes four of the five suggestions.

Prior findings

Finding Status Evidence
Should 1: ruled candidate us-nj:medicaid-caretaker-relative-any-adult had no linked ruling record Fixed Details below the table. No test can pin a ruling link, so the mutation rule doesn't apply here; I verified it by reading the record.
Suggestion 1: pe_us.prs: [] Fixed Lists https://github.com/PolicyEngine/policyengine-us/pull/10984. The URL matches the guard's prs pattern.
Suggestion 2: no New Jersey negative student case Fixed Case 4 (NJ parent with an 18-year-old who is not a student) expects [false, false]. Dropping is_student from the student branch flips the parent to true, so Case 4 pins that term for New Jersey. rulespec-us#2261 now uses it as companion row 4.
Suggestion 3: pin the reported medicaid_category symptom Fixed Case 1 asserts medicaid_category: [PARENT, OLDER_CHILD]. The hand trace is below. With NJ any_adult_option set to false, the adult becomes ADULT and the case fails.
Suggestion 4: source other states' S25 pages Declined, accepted The PR body says "Other states' S25 pages stay deferred to #10775". The 50 false values are still labelled "not yet researched" in the parameter. This was optional.
Suggestion 5: NJ requires_deprivation started 2018 although S25 removes deprivation from 2014 Fixed NJ is now 2014-01-01: false, with an S25 #page=31 reference and a quoted comment. Checked against the page below. rulespec-us#2261 required output 3 now cites the 2014 PE-US value.

Should 1, in detail. The candidate now carries ruling: {by: PolicyEngine, date: '2026-11-15', record: https://github.com/PolicyEngine/policyengine-us/pull/10984#issuecomment-6927855962}, and its note opens "Ruled 2026-11-15 by the project lead; the ruling record is …".

New findings

Critical

None.

Should address

None.

Suggestions

None.

Verified

  • The candidate at head has every field CONTRIBUTING.md#source-conflicts step 1 requires.
    • id, program: medicaid and question.
    • years (from 2014, calendar).
    • Four sources: plan, two regulations, and the new N.J.A.C. 10:78 regulation. All are https:// and all kinds are in SOURCE_KINDS.
    • readings and example.
    • pe_us with follows: statute (a valid value), two code paths that are tracked files at head and in the merged tree, and prs.
    • state: ruled.
    • There is no marker (step 1). The id is not in the vendored register, and the merged tree holds exactly one entry with this id.
  • The new N.J.A.C. 10:78-1.1(a) source. PDF page 4 of the nj.gov compilation (current through 57 N.J.R., June 16, 2025) reads "criteria for NJ FamilyCare eligibility for certain uninsured parents, caretakers and caretaker relatives and children not eligible under the provisions of N.J.A.C. 10:69". The #page=4 anchor and the note's quote are correct.
  • The S25 deprivation option. NJ-13-0011-MM1, PDF page 31 (scan; approval 02/28/2014, effective 01/01/2014), under "Options relating to the definition of dependent child", selects "The state elects to eliminate the requirement that a dependent child must be deprived of parental support or care by reason of the death, physical or mental incapacity, or absence from the home or unemployment of at least one parent."
    • The new comment in requires_deprivation.yaml quotes this correctly.
    • The new reference follows the CLAUDE.md rule for a single cited page: the page is in the href only. Its title style matches the existing S25 references in student_option.yaml and any_adult_option.yaml.
    • Reach. Every other state's value starts 2018-01-01, and NJ income_limit starts 2018-01-01. Backdating NJ therefore changes no result from 2018 on, and only fills in New Jersey for 2014 to 2017.
  • Hand-checks of the new and changed expectations (2026):
    1. Case 1, adult medicaid_category = PARENT.
      • The adult is not an SSI recipient, is outside the infant, young-child and older-child age ranges (40 ≥ 19), and is not pregnant.
      • is_parent_for_medicaid = nfc (true, as in round 1) and fc. For fc, medicaid_parent_income_limit for NJ is income_limit["NJ"] = 0.34 (from 2021), and $3,600 / $21,640 = 16.6%, which is under 0.34.
      • The parent group comes before ADULT in medicaid_category's order, so the adult is PARENT.
    2. Case 1, child medicaid_category = OLDER_CHILD.
      • Age 8 is in the older-child range: 6 is true, 19 is false.
      • NJ older-child income_limit is 1.47, and 16.6% is under it.
      • The earlier categories don't apply: the child is not an SSI recipient and not under age 6.
    3. Case 1, mutation. With NJ any_adult_option set to false, the foster child is excluded and the adult's nfc is false. The adult then falls to ADULT, since 16.6% is under the 138% expansion limit, so the assertion fails. Main gives ADULT here, which matches the partner's report.
    4. Case 4, parent = false.
      • The 18-year-old fails age < age_limit["NJ"] (18 < 18).
      • is_full_time_student = is_full_time_college_student (input, default false) or is_in_k12_school (ages 5 to 17 only), so it is false at 18. The student branch is therefore false.
      • With no dependent child in the tax unit, meets_basic_criteria is false.
    5. Case 4, 18-year-old = false. The 18-year-old is a tax-unit dependent, so ~is_dependent fails.
    6. Case 4, mutation. Removing is_student & from the student branch makes age_qualifies true (NJ elects the option, and 18 < 19), so the parent becomes true and the case fails.
    7. PR-body table. On main, NJ student_option was false, so Case 4 is false on both main and this PR, as the table says.
  • Unchanged since round 1. The fix commit leaves is_parent_for_medicaid_nfc.py, any_adult_option.yaml, student_option.yaml and the changelog fragment untouched. Cases 1 to 3 keep their inputs and parent expectations, so the round-1 hand-checks and mutation checks still hold.
  • Source-conflict ratchet (Tie source-conflict markers to the register shared with Axiom #10287) on the merged tree.
  • The NH Add New Hampshire state supplementation for 2026 #10156 pattern is still clear. Keys come from state_code_str, and both indexed parameters keep all 51 keys.

Axiom

axiom: TheAxiomFoundation/rulespec-us#2261 queued is valid.

  • Status. The issue is OPEN and labelled pe-parity. It was updated on 2026-11-16 at 20:16 UTC, after the fix commit.
  • It matches the head. It cites head 29cbb14acd and links the test file at that commit.
  • The rows changed in this round:
    • Companion row 4 is now Case 4 (a NJ 18-year-old who is not in school gives caretaker = false, from 435.4 "Dependent child" (1) and S25).
    • Row 1 notes Case 1's medicaid_category PARENT assertion.
    • Required output 3 says deprivation is eliminated from January 1, 2014, citing PE-US's 2014-01-01 value.
  • Still dispatch-ready. The issue has the module paths, corpus citations, verbatim 435.4 and S25 text, readings including the ruled plan-over-10:69 conflict, required outputs and companion tests from 435.4 and S25. Its blocker is still recorded: the pinned corpus release has no New Jersey Medicaid state plan scope.

CI

All 37 checks pass at head 29cbb14acd (run created 2026-11-16 at 20:14 UTC). They include:

  • PR validation (lock, guard tests, lint, changelog);
  • Household API Partners;
  • every baseline shard, including contrib-hhs;
  • every contrib shard;
  • the Python shards and Microsimulation.

I re-checked at the end of the review. Nothing is pending or failing.

Branch

Verdict: APPROVE

@DTrim99 DTrim99 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.

Approving (round 2): the prior should-address item is fixed (the candidate links the dated ruling record), and suggestions 1, 2, 3 and 5 were taken (4 deferred to #10775). CI 37/37 green at 29cbb14. Full report in the comment above.

@hua7450
hua7450 merged commit f6c353c into main Nov 17, 2026
37 checks passed
@hua7450
hua7450 deleted the nj-medicaid-caretaker-definition branch November 17, 2026 15:37
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants