MS-DRG Grouper Overview¶

The MS-DRG (Medicare Severity Diagnosis Related Groups) grouper classifies patient claims into diagnosis-related groups based on diagnoses, procedures, demographics, and other clinical factors.

How it works¶

The grouper follows the CMS DRG classification logic:

graph LR
    A[Input Claim] --> B[Preprocessing]
    B --> C[Grouping]
    C --> D[HAC Processing]
    D --> E[Final DRG]

    B -->|"Load attributes,<br>apply exclusions"| C
    C -->|"Evaluate DRG<br>formulas"| D
    D -->|"Hospital-Acquired<br>Condition rules"| E

1. Preprocessing — loads diagnosis attributes, applies exclusions, clusters codes, sorts by clinical significance
2. Grouping — evaluates DRG formulas against the claim's clinical profile
3. HAC Processing — applies Hospital-Acquired Condition rules based on hospital_status
4. Final Grouping — produces the final DRG assignment

The entire pipeline mirrors the CMS Java grouper and is validated claim-by-claim against it.

Basic usage¶

import msdrg

with msdrg.MsdrgGrouper() as g:
    result = g.group({
        "version": 431,
        "age": 65,
        "sex": 0,
        "discharge_status": 1,
        "pdx": {"code": "I5020"},
    })

Helper function¶

The create_claim() convenience function builds the nested dict structure for you:

claim = msdrg.create_claim(
    version=431, age=65, sex=0, discharge_status=1,
    pdx="I5020", sdx=["E1165"], procedures=["02703DZ"]
)

See the full function signature in the API Reference.

Thread safety¶

The grouper context is immutable after initialization and safe to share across threads. Each call to group() is thread-safe.