Medicare Spending Per Beneficiary — Hospital

Overview

The Medicare Spending Per Beneficiary (MSPB) — Hospital dataset is published by CMS as part of the Hospital Compare provider data files on data.cms.gov (dataset identifier 5hk7-b79m). It contains hospital-level spending data for the MSPB measure, which captures total Medicare Part A and Part B spending for episodes of care spanning 3 days prior to an index hospital admission through 30 days after discharge. This spending window encompasses inpatient costs, post-acute care (skilled nursing facility stays, home health services, inpatient rehabilitation), and all outpatient services billed by any provider during the episode — not just the admitting hospital. The current file covers the FY2026 reporting period.

MSPB is expressed as a ratio to the national median episode cost. A value of 1.0 means a hospital's per-episode spending equals the national median; values above 1.0 indicate higher spending. Most hospitals cluster between 0.85 and 1.15. The measure is price-standardized to remove geographic wage index differences, isolating utilization intensity rather than regional price variation. This dataset answers questions such as: which hospitals generate higher total episode costs, how much of a hospital's spending is driven by post-acute care referral patterns, and how a hospital's resource utilization compares to the national median after removing price effects.

Join Strategy

This dataset joins to hospital entity pages on CareGraph using the Facility ID field, which contains the CMS Certification Number (CCN) as a 6-digit zero-padded string. During ETL, the normalize_ccn() function strips whitespace and zero-pads values shorter than 6 characters to ensure consistent matching. The _load_mspb() function reads the source CSV column using a candidate-list strategy that checks for "Facility ID", "Hospital CCN", "Provider Number", "Facility Id", and "CCN" to accommodate CMS header variations across vintages. Unlike multi-measure datasets (e.g., HRRP), MSPB produces a single record per hospital rather than an array of condition-level rows. The ETL attaches this record to the hospital manifest at data.mspb as a flat dictionary. The join is a left join from the hospital manifest to the MSPB data — hospitals without MSPB records display missing data indicators rather than being excluded from the site.

Known Limitations

• Suppression for low volume. Hospitals with fewer than 25 eligible episodes are excluded from the MSPB measure and appear with suppressed values. This disproportionately affects small and rural hospitals and is suppression, not an indicator of low spending.
• Medicare fee-for-service only. MSPB covers only Original Medicare (fee-for-service) beneficiaries. Medicare Advantage enrollees, Medicaid-only patients, commercially insured patients, and uninsured patients are excluded. In markets with high MA penetration, the MSPB ratio is calculated on a smaller, potentially non-representative subset of a hospital's total episodes.
• No case-mix adjustment beyond DRG. The measure is price-standardized to remove geographic wage index differences but does not adjust for patient severity or complexity beyond the DRG assignment. Hospitals treating sicker patients within the same DRG will show higher MSPB ratios even if their care is clinically appropriate.
• Post-acute spending attribution. All spending during the episode window is attributed to the hospital that performed the index admission, including costs billed by skilled nursing facilities, home health agencies, and outpatient providers. Hospitals that discharge to expensive post-acute settings will show elevated MSPB even if their own inpatient costs are low. This attribution model rewards hospitals that discharge patients to lower-cost settings regardless of clinical appropriateness.
• Measurement lag. The FY2026 data uses a multi-year measurement window that lags the reporting year. Recent changes in a hospital's discharge patterns or post-acute referral networks are not reflected in the current MSPB values.
• IPPS hospitals only. The measure applies to hospitals paid under the Inpatient Prospective Payment System. Critical Access Hospitals, Maryland waiver hospitals, and hospitals in US territories may not appear or may not be comparable.

Data Quality Notes

• Numeric fields stored as strings. The source CSV encodes MSPB ratio and spending values as strings. Suppressed rows contain "Not Available" or "Not Applicable" instead of numeric values. The ETL parses these with _try_float(), converting valid numbers to floats and discarding sentinel strings so they appear as null in the JSON manifest.
• Column name variation across vintages. CMS has changed column header casing and naming between file releases (e.g., "Facility ID" vs. "Facility Id"). The ETL uses a candidate-list column matching strategy in _load_mspb() that checks multiple known column names to handle these variations without manual mapping updates.
• Single-record-per-hospital structure. Unlike other Hospital Compare datasets that have multiple rows per hospital (one per measure), MSPB produces one row per hospital. The ETL stores this as a flat dictionary rather than an array, so consumers should expect data.mspb to be an object, not a list.
• Missing value normalization. The source data uses "Not Available", "Not Applicable", and empty strings for missing values. The ETL filters all of these out during the _clean() step, so any field not present in the manifest record was either missing or suppressed in the source.