National and State Healthcare-associated Infections Standardized Infection Ratio Report

Using Data Reported to the National Healthcare Safety Network

January – December 2010

Methods

Eligible Data

This report includes data from surveillance activities performed during 2010 and reported either mandatorily or voluntarily by healthcare facilities to NHSN from facilities across all 50 states, Washington, D.C., and Puerto Rico. Data used in these calculations were restricted to the most up-to-date NHSN definition for CLABSI in 2008,⁶ and for CAUTI in 2009.⁷ Any data reported from non-acute care hospitals (e.g., long-term care hospitals) and from dialysis wards or facilities were excluded from this report. All acute care hospital patient care locations were included in this report. Data were accessed October 5, 2011, to allow for a 10-month latency period to help ensure complete reporting of HAIs and denominator data through December 2010. This is slightly extended compared to the 6-month latency period of previous reports.

To illustrate the degree to which facilities reported to NHSN during 2009 and 2010 in the United States, this report presents the number of facilities and the number of patient care locations reporting within each state. In addition to presenting an all-inclusive category of locations, locations were also stratified into three mutually exclusive groups that reported to NHSN, by state: (for this report, wards also include step-down, specialty care areas ^{including hematology/oncology, bone marrow transplant}, and long-term acute care locations/facilities), and neonatal intensive care units (includes Level II/III and Level III). To facilitate an assessment of growing capacity for NHSN to be used for surveillance and prevention activities, these reporting characteristics have been summarized for each reporting period (2009 and 2010). Although comparisons of HAI experience were made at the location level using existing descriptions of location, further work is needed to confirm the accurate categorization of several location types by reporting facilities. These types include improved categorization of long-term acute care locations and confirmation of accurate mapping of hematology/oncology locations. Therefore, summary statistics for these distinct patient-care areas will be considered in future years after confirmation and accurate categorization has occurred.

The SSI data reported here include only a subset of the operative procedures on which facilities perform surveillance and report HAI data. This subset includes many of the more commonly reported procedures and approximates those targeted for process-of-care improvements by the Surgical Care Improvement Project (SCIP), a national program led by CMS and CMS-funded Quality Improvement Organizations.⁸ CDC compared these procedures to NHSN procedure categories and determined the most appropriate mapping between the two groups of procedures. In several instances, multiple NHSN procedure codes were mapped to a single SCIP procedure (Appendix A). This list of procedures is the same as those procedures specified in the HHS Action Plan as targets of SSI prevention.⁷ However, there are notable differences between procedure groupings included in SCIP and those in this report. These include, but are not limited to, inclusion of both primary and revision hip arthroplasties in the NHSN hip arthroplasty procedure category (while only primary hip arthroplasty is included in the CMS SCIP grouping).

SSI SIRs were reported for the aggregate across all of these procedure types, as well as for each specific procedure category. Consistent with the HHS Action Plan, CDC further limits the SSIs included in this report to a subset of all SSIs reported as deep incisional and organ/space infections that were detected during the hospital admission where the operation was performed or upon readmission to that same hospital.⁹ Superficial incisional SSIs and any SSIs identified through post-discharge surveillance were excluded in alignment with current recommendations for public reporting summary measures.¹⁰

Basic summary statistics of characteristics of reporting by hospitals are presented for each state and nationally. Data external to NHSN were required to compile some of these metrics. Specifically, CDC consulted with each state health department to verify the number of acute care facilities eligible for reporting to NHSN, date of implementation of any mandated reporting, and the performance of any external validation of the reported data. Validation included any data quality assessment of missing or implausible values and/ or detection of outlier facilities (e.g., number of infections, rates, denominators), and/or audits of medical records. Information on validation efforts was requested from all states, regardless of presence of a legislative mandate for the particular HAI type. Some states without mandatory reporting of a given HAI have performed validation on NHSN data that is voluntarily shared with them by facilities.

Top of page

Summary HAI Data and Calculation of SIRs

The referent period remained January 2006 through December 2008, as in previous SIR reports.⁴ However, for CAUTI, the referent period is 2009. All facilities reporting at least 1 month of relevant data to NHSN during the referent time period (regardless of any mandate) were included in the referent period; these data are comparable to those reported in the NHSN annual report.^1,5 The SIRs represent comparisons of observed HAI occurrence during each distinct reporting period with the predicted occurrence based on the rates of infections among all facilities adjusting for key covariates (referent population). Illustrative examples of how an SIR is calculated are provided in previous reports (http://www.cdc.gov/HAI/surveillance/statesummary.html).⁴ Although over 40 patient locations are represented in the referent time period,1 during subsequent years, such as 2010, some facilities reported HAI data from location types not represented in the referent time period. In these scenarios, an SIR cannot include data from these new location types.

The CLABSI and CAUTI SIRs are adjusted for patient mix by type of patient care location, hospital affiliation with a medical school, and bed size of the patient care location. Other factors, such as hospital bed size, were not consistently associated with differences in CLABSI or CAUTI rates after accounting for patient location and, therefore, were not included in CLABSI SIR risk adjustment. For NICUs, the pooled mean umbilical catheter-associated BSI (UCAB) rate and the CLABSI infection rate within each of the five birth weight categories were used to determine the predicted number of device-associated BSIs from each reporting facility, referred to as CLABSIs for this report.¹ Of note, clinical sepsis (without laboratory-confirmed bloodstream infection) was not included in the calculations of CLABSI during either the reporting period or referent period. CAUTIs are not reported from NICUs.

For SSI SIRs, risk models were constructed evaluating all available procedure-related risk factors (e.g., duration of surgery, surgical wound class, use of endoscopes, status as re-operation, patient age, and patient assessment at time of anesthesiology ^{ASA score}) to provide the best possible adjustment for differences in patient-mix within each type of surgery. These risk models were constructed specifically for this report to predict SSIs reported as deep incisional or organ/space infections and only those detected during admission or upon readmission to the same hospital.² A summary of the adjustment parameters are listed in Appendix A. As additional procedure-specific data become available to NHSN improved risk models can be constructed.

For the national and state SIR, all eligible data were included and the total number of infections predicted was compared to the number observed at each level of aggregation. Second, facility-specific SIRs were also calculated for each of the summary measures presented nationally. However, if a single facility’s predicted number of HAIs (e.g., CLABSI) was <1.0, a facility-specific SIR was neither calculated nor included in the determinations of the distribution of facility-specific SIRs. This report considered calculations of a facility-specific SIR as reliable only when at least one HAI would be predicted based on the data reported to NHSN from that facility. For the state-specific aggregation of HAI data, state-specific SIRs were not reported unless at least five facilities reported data. In addition, if fewer than 20 facilities had reliable facility-specific SIRs, then no key percentile distributions of facility-specific SIRs were calculated (such as occurred with abdominal aortic aneurysm repair at the national level) for that level of reporting. Because most states had sufficient reporting from facilities to calculate reliable state-specific SIRs, including roughly half with sufficient reporting to reliably calculate key percentile distributions, state-specific summary data is reported for all states.

An SIR of 1.0 should be interpreted as indicating that the number of events the entity (e.g., state health department, healthcare facility) observed is no different than if its experience had been the same as that of the referent population. Because the SIR is an estimate based on calculations of reported data, confidence intervals (CIs) are calculated to allow for accurate interpretation of the SIR. If these CIs include a value of 1.0, the SIR should be interpreted as if it were 1.0. The CI around the SIR depends on several factors, including the number of facilities reporting data from the relevant patient care locations, the number of device days or operative procedures that were reported, and the types of facilities reporting.

Top of page

Serial Comparison of SIRs

The evaluation of progress in the prevention of HAIs was assessed by comparing the SIRs between sequential years. This was first accomplished by comparing the SIRs between each of the sequential reporting periods. A second sensitivity analysis was performed by restricting the reporting facilities to only those that reported during the initial reporting period, referred to as the change in SIR for continuously reporting facilities. A conditional binomial test was performed to assess statistically significant changes in the pairs of sequential SIRs (two-sided P-value ≤.05). If the change was not statistically significant, it was reported as "no change." Prevention success can be measured as sustained (similar SIRs between reporting periods), improved (SIRs sequentially decreasing), or slowing (SIRs sequentially increasing toward or above 1.0). Because data for this report uses all data reported to NHSN through October 5, 2011, calculations of the 2009 SIR will differ slightly from reports using datasets created earlier in time, including those reported by individual state health departments.

Top of page