Validation

CMS TEAM Analyzer is designed to run in your dbt project alongside The Tuva Project. Validation should confirm that your Tuva inputs are complete enough for TEAM analytics, that package configuration matches your reporting periods, and that the published marts reconcile to known source-data totals.

Recommended dbt Commands

After CMS TEAM Analyzer is added to packages.yml, run the package in the same dbt project where Tuva is already installed:

dbt deps
dbt seed --full-refresh --select package:cms_team_analyzer
dbt run --select +package:cms_team_analyzer
dbt test --select package:cms_team_analyzer

If you load participant-specific CMS target prices into your own model or source table, include that model in your normal upstream build before running the TEAM package.

What The Package Tests

Package tests focus on model-level integrity:

Test area	Purpose
Episode identity	Confirms each published episode has a stable, non-null episode_id
Overview snapshots	Confirms reporting snapshots are present
Performance periods	Confirms monthly performance rows have a valid period month
Drivers and opportunities	Confirms driver and opportunity rows have required identifiers
Target-price resolver	Confirms target-price source metadata is present where expected

What Your Project Should Validate

Before using the marts for operational reporting, confirm:

• Tuva source models include the required claim, encounter, eligibility, patient, practitioner, and terminology fields.
• Performance-year and baseline-year vars match your reporting intent.
• team_spend_basis is appropriate for your source data.
• Participant-specific CMS target prices are loaded when available.
• Optional enhanced fields are present where you expect higher CMS-methodology fidelity.
• mart_team__episode episode counts and spend totals reconcile to trusted samples.
• High-dollar sample episodes match finance or reimbursement-team expectations.

Suggested Implementation Checks

Check	Why it matters
Compare source claim paid totals to TEAM assigned spend	Confirms claim assignment is not unexpectedly dropping or duplicating spend
Review a sample of LEJR and Spinal Fusion outpatient anchors	Confirms outpatient HCPCS triggers and outpatient-to-inpatient precedence behave as expected
Review excluded readmission examples	Confirms excluded MS-DRG and MDC logic matches your interpretation of CMS policy
Review open episodes	Confirms snapshot-date logic and incomplete runoff assumptions are clear to business users
Compare CMS public regional targets to participant-specific target files	Confirms the target-price precedence is working as intended

Development Test Project

Illuminate maintains an internal dbt test project with synthetic CCLF-shaped data to exercise package behavior before release. That project is for package development and regression testing. Client implementation docs intentionally focus on your dbt project because that is where Tuva, CMS TEAM Analyzer, and your production source data run together.