homelab/stack
1
0
Fork 0
Code Issues 57 Pull Requests 1 Actions Packages Projects 1 Releases Wiki Activity

update notebook to use staging steps instead of SQL hack

Browse Source 
Remove the inline core.medical_claim → core.condition SQL JOIN from
load() — the pipeline's _stg_medical_claim and _stg_medical_claim_condition
steps handle diagnosis code enrichment now. Add staging steps to the
UAMCC pipeline execution and walkthrough sections.
This commit is contained in:
kert
2026-03-07 15:05:45 -05:00
parent b2a7a0b037
commit 2b5e2780a8
1 changed files with 4 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
notebooks/cms_quality_measures.py
View File

@@ -138,6 +138,8 @@ def _(ex, inspect, mo):
@app.cell(hide_code=True) @app.cell(hide_code=True)
def _(ex, fn_doc, mo): def _(ex, fn_doc, mo):
_uamcc_steps = [ _uamcc_steps = [
("0a. Stage Medical Claims", ex.stg_medical_claim),
("0b. Stage Claim–Condition Pairs", ex.stg_medical_claim_condition),
("1. Performance Period", ex.uamcc_performance_period), ("1. Performance Period", ex.uamcc_performance_period),
("2. MCC Cohort Identification", ex.uamcc_int_mcc_cohort), ("2. MCC Cohort Identification", ex.uamcc_int_mcc_cohort),
("3. Denominator", ex.uamcc_int_denominator), ("3. Denominator", ex.uamcc_int_denominator),
@@ -657,18 +659,6 @@ def _(con, inspect, mo, pl):
def load(table_ref): def load(table_ref):
"""Load a table from DuckDB, with fallbacks for missing tables.""" """Load a table from DuckDB, with fallbacks for missing tables."""
# Enrich core.medical_claim with diagnosis codes from core.condition
if table_ref == "core.medical_claim":
return con.execute("""
SELECT m.*,
c.normalized_code,
c.normalized_code AS principal_diagnosis_code
FROM core.medical_claim m
LEFT JOIN core.condition c
ON m.claim_id = c.claim_id
AND c.condition_rank = 1
""").pl()
_schema, _tbl = table_ref.split(".", 1) _schema, _tbl = table_ref.split(".", 1)
try: try:
_df = con.execute(f'SELECT * FROM "{_schema}"."{_tbl}"').pl() _df = con.execute(f'SELECT * FROM "{_schema}"."{_tbl}"').pl()
@@ -724,6 +714,8 @@ Running all three measure pipelines against **Synthea/Tuva claims data**
@app.cell(hide_code=True) @app.cell(hide_code=True)
def _(ex, mo, pl, run_step, uamcc_pp): def _(ex, mo, pl, run_step, uamcc_pp):
_uamcc_pipeline = [ _uamcc_pipeline = [
("cms_quality_measures._stg_medical_claim", ex.stg_medical_claim),
("cms_quality_measures._stg_medical_claim_condition", ex.stg_medical_claim_condition),
("cms_quality_measures._uamcc_performance_period", ex.uamcc_performance_period), ("cms_quality_measures._uamcc_performance_period", ex.uamcc_performance_period),
("cms_quality_measures._uamcc_int_mcc_cohort", ex.uamcc_int_mcc_cohort), ("cms_quality_measures._uamcc_int_mcc_cohort", ex.uamcc_int_mcc_cohort),
("cms_quality_measures._uamcc_int_denominator", ex.uamcc_int_denominator), ("cms_quality_measures._uamcc_int_denominator", ex.uamcc_int_denominator),