Program Schedule

NHSN vs. CMS: How Changing the Risk Adjustment Model Changes the Results

Session: Oral Abstract Session: Advancing Public Reporting and Surveillance of Healthcare-associated Infections
Thursday, October 9, 2014: 10:45 AM
Room: The Pennsylvania Convention Center: 109-AB

NHSN vs. CMS:  How Changing the Risk Adjustment Model Changes the Results

Background: In 2012, the Centers for Medicare and Medicaid Services (CMS) made it mandatory for all hospitals participating in the Inpatient Prospective Payment System (IPPS) to report surgical site infections (SSI) associated with qualifying abdominal hysterectomy (HYST) and colon procedures (COLO) to the National Healthcare Safety Network (NHSN). These data are then made available to the public on Medicare's Hospital Compare website.

Also in 2012, NHSN began using an improved complex risk adjustment model for SSI. Despite NHSN's improved methodology, CMS instead uses a different, simple risk adjustment model for the publicly reported data. Figure 1 details the differences between the two datasets. The objective of this study was to determine if there was a difference in the reported standardized infection ratio (SIR) using the two different models.

Methods: Using 2013 HYST and COLO NHSN data, SIRs were calculated for 10 acute care hospitals in a large healthcare system using two output reports following the criteria in Figure 1: “SIR - In-plan Complex AR SSI data by procedure”; and “SIR - Complex 30-Day SSI Data for CMS IPPS”. Facilities with an expected infection of less than 1 in either category were excluded. The resulting SIRs for the 4 remaining facilities were compared.

Results: See Figure 2. Changes in SIR ranged from -12% to +31% for COLO and -24% to -9% for HYST. None of these differences were statistically significant.

Conclusion: Despite there being no statistical differences between the two analysis methods, differences in reported SIRs can cause confusion when reviewing a facility's data. NHSN's updated risk model was created in order to more fairly compare facilities against the aggregate population. By ignoring this method, CMS has created a separate model that must be explained by Infection Preventionists to stakeholders and hospital leadership. Given that the SIRs are not statistically different, it would reduce confusion to analyze the data using the same model in both systems.

Kathleen Gase, MPH, CIC1, Raya R. Khoury, MPH1 and Hilary M. Babcock, MD, MPH2, (1)BJC Healthcare, St. Louis, MO, (2)Division of Infectious Diseases, Washington University School of Medicine, St. Louis, MO


K. Gase, None

R. R. Khoury, None

H. M. Babcock, None

Findings in the abstracts are embargoed until 12:01 a.m. EDT, Oct. 8th with the exception of research findings presented at the IDWeek press conferences.

Sponsoring Societies:

© 2014, All Rights Reserved.

Follow IDWeek