Unplanned Hospital Visits — Hospital

Overview

The Unplanned Hospital Visits — Hospital dataset is published by the Centers for Medicare & Medicaid Services (CMS) as part of the Hospital Compare program, now integrated into the Care Compare initiative on data.cms.gov (Provider Data API identifier 632h-zaca). It contains hospital-level measures for unplanned return visits following outpatient procedures, including ED visits after hospital outpatient surgery (OP-36) and unplanned readmissions after outpatient procedures. Each row represents one measure for one hospital, with fields for the observed rate, national comparison category, number of patients, and measure start/end dates. The current file covers FY2026, using a measurement window of approximately 3 years ending in 2022.

This dataset complements the Hospital Readmissions Reduction Program (HRRP) measures by capturing outpatient-to-ED and outpatient-to-inpatient transitions that HRRP's inpatient-focused readmission measures do not cover. It answers questions such as: which hospitals have higher- or lower-than-expected rates of unplanned return visits after outpatient surgery, how a hospital's ED return rate compares to the national average, and whether outpatient procedures at a given facility lead to subsequent emergency or inpatient care at rates that differ from national benchmarks. The data covers Medicare fee-for-service claims from IPPS-participating acute care hospitals across the United States.

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 (e.g., 010001). During ETL, the normalize_ccn() function strips whitespace and zero-pads values shorter than 6 characters to ensure consistent matching. The generic _load_measures_by_ccn() loader reads the source CSV, identifies the CCN column using a candidate-list strategy (checking "Facility ID", "Hospital CCN", "Provider Number", and variants via _find_column()), and groups all measure rows by normalized CCN. Each hospital's measure-level records are attached to its JSON manifest under the unplanned_visits key as an array of per-measure objects. Non-numeric values such as "Not Available" are filtered out during loading; numeric fields are parsed via _try_float(). The join is a left join from the hospital manifest — hospitals without Unplanned Hospital Visits records retain their existing data and display missing indicators for this dataset.

Known Limitations

• Small-volume suppression. Hospitals with fewer than 25 cases in the denominator for a given measure have their rates suppressed, displayed as "Too Few to Report." This disproportionately affects small and rural hospitals, and suppressed entries should not be interpreted as zero unplanned visits. Suppressed rows are converted to null in the JSON manifest.
• Medicare fee-for-service only. The dataset covers only Original Medicare (fee-for-service) claims. Medicare Advantage enrollees, Medicaid-only patients, commercially insured patients, and the uninsured are excluded. In markets with high Medicare Advantage penetration, reported rates are calculated on a smaller, potentially non-representative subset of the hospital's total outpatient surgical volume.
• Observation stay classification. Observation stays (outpatient status) are counted differently than inpatient admissions. Whether a return visit is classified as an "ED visit" or a "readmission" depends on the patient's admission status, not the clinical severity of the event. This can misrepresent the nature and acuity of return visits, and undercount true return rates when patients are placed in observation status rather than formally admitted.
• Incomplete procedure-level risk adjustment. Hospitals that perform mostly low-risk outpatient procedures will naturally have lower return rates than hospitals performing complex outpatient surgeries. The risk adjustment does not fully account for procedure-level complexity differences, which penalizes hospitals with a more complex outpatient surgical mix.
• Planned readmission algorithm misclassification. The measures use CMS's planned readmission algorithm to identify "unplanned" visits from claims data. Some clinically planned follow-ups may be misclassified as unplanned if the billing codes do not match the algorithm's expected patterns, inflating unplanned visit rates for hospitals with atypical coding practices.
• 3-year measurement lag. The measurement window ends approximately in 2022 for the FY2026 release. Recent changes in a hospital's outpatient surgical practices or quality improvements are not reflected in the current data. The window also overlaps with the COVID-19 public health emergency, during which outpatient surgical volumes and case mix were disrupted.

Data Quality Notes

• Rate and score fields stored as strings. The source CSV encodes observed rates, denominators, and comparison fields as string values. Suppressed rows contain "Too Few to Report" or "Not Available" instead of numeric values. The ETL parses these with _try_float(), converting non-numeric entries to null in the JSON manifest. Rows where the cleaned value equals "Not Available" or "Not Applicable" are excluded entirely during loading.
• Column name variation across vintages. CMS changes column header casing and naming between file releases (e.g., "Facility ID" vs. "Facility Id", "Measure ID" vs. "Measure Name"). The ETL uses a candidate-list column matching strategy via _find_column() to handle these variations without manual updates.
• Missing value encoding inconsistency. The source data uses "Not Available", "N/A", empty strings, ".", and "*" interchangeably for missing values depending on the field. The ETL normalizes all of these to null via the _try_float() and _clean() functions. Measure start and end date fields use MM/DD/YYYY string format and are not converted during ETL.
• Claims-derived measure definitions. All measures are derived from administrative claims data, not clinical chart review. Variation in hospital coding practices — particularly how outpatient procedures and subsequent ED visits are billed — can affect reported rates independently of actual clinical outcomes. Hospitals with more thorough or atypical coding may show systematically different rates.