National and State Healthcare-associated Infections Standardized Infection Ratio Report

Using Data Reported to the National Healthcare Safety Network

January – December 2011

Discussion

The HAI data summarized in this report demonstrate that healthcare facilities reporting to NHSN during 2011, as a group, reported fewer CLABSIs (41%), CAUTIs (7%), and SSIs (17%) than predicted based on the case-mix of patients and locations that were monitored. Moreover, CLABSI prevention success improved between reporting periods, as the SIR during 2011 was significantly lower compared to the previous year (2010: SIR 0.68, 32% reduction). Improved prevention success was evident in all location groups (critical care, ward, and NICU) for CLABSI. Improvement was more modest for SSIs, for which the overall SSI SIR decreased from 0.93 to 0.83, but was not evident for all of the procedure types and only for knee arthroplasty when limited to continuous reporters. Of note, in 2011 a substantial proportion of all procedures included in this report were reported by facilities in California as they began to report for the first time in response to a state-wide mandate. Measuring progress between the two years therefore may be better assessed by focusing on the continuous reporters. The experience in CAUTI prevention is less clear. Although there were modest reductions in the SIR between 2010 and 2011, the decrease was driven by the 550 facilities reporting CAUTI from wards during both 2010 and 2011. In contrast, there was essentially no significant difference in the SIRs in critical care locations between the two periods. The lack of significant reductions in CAUTI SIRs may be due to lack of substantial progress in critical care areas, an inability to substantially decrease catheter days in critical care areas (as can be done more easily in wards), or both of these factors. However, at least one state, Michigan, has seen a 25% reduction in CAUTI using a device-day rate based SIR after implementing a series of prevention initiatives. This suggests that with continued prevention efforts, we should expect continued reductions in both critical care and ward-specific CAUTI SIRs using a device-day methodology as described in this report (13).

This SIR report is the first to provide some perspective on the potential improvements that can occur with facility-specific engagement. For each major location group and procedure category, roughly 2-9% of the facilities reported SIRs significantly greater than 1.0. Although the specific number of facilities represented by this group varies between HAI type and procedure (e.g., 54 facilities for CLABSI, 133 for CAUTI, 25 for hip arthroplasty, 30 for knee arthroplasty, 20 for colon surgery, and 15 for abdominal hysterectomy), it is a relatively small number of facilities compared to total number of facilities reporting in 2011 (e.g., 3,468 reporting CLABSI, 1,802 reporting CAUTI, 2,130 reporting SSIs). Focusing efforts on these outlier facilities may be one strategy to focus prevention resources in coming years, although most efficient methods to target prevention activities to make substantial reductions nationally are still being explored.

Overall compared to the previous year, there was an increase of about 1,200 facilities reporting CLABSI, 900 facilities reporting CAUTI, and 700 facilities reporting SSI. This dramatic increase is mostly the result of new reporting requirements for hospitals participating in CMS’s Hospital IQR Program, requiring participating facilities to report to CMS, through NHSN, ICU CLABSIs starting in 2011 and CAUTI and SSI beginning in 2012(1). Summary data reported through NHSN to CMS as part of this program and posted quarterly on the CMS Hospital Compare website are a subset of the data reported here (some facilities report to NHSN but do not participate in the IQR Program); therefore the summary statistics are expected to vary slightly.

Using the most recent data available in NHSN, we estimated 12,400 CLABSIs (95% CI, 11,500-13,300) occurred among critical care patients in 2011; the estimated total number of superficial incisional, deep incisional, or organ/space SSIs that occurred in 2011 (among the estimated 3,011,412 surgical procedures evaluated) was 52,567 (45,332-60,844). These infections cost CMS approximately $26,000 per CLABSI occurring in ICU patients. The attributable reimbursement from SSIs has not been determined to date. Also, because the distribution of major payer categories (i.e. Medicare, private insurance, and Medicaid) among patients with CLABSIs is unknown, we could not estimate the number of infections and total reimbursements attributable to these infections separately by major payer categories. While approximately 39% of all hospital costs result from care to Medicare beneficiaries, another 16% result from care provided under Medicaid and 35% from beneficiaries of private payers (14). Meanwhile, the per-infection reimbursements from private payers are likely to be considerably higher than that from Medicare and Medicaid (15). Thus, simply multiplying the point estimate of the burden of ICU CLABSIs by the attributable Medicare reimbursement per infection, while equaling approximately $322 million, likely underestimates the national total reimbursements attributable to these infections and borne by all third party payers.

Regarding CLABSI prevention success regionally, almost all of the jurisdictions with sufficient data had overall CLABSI SIRs significantly less than 1.0 in 2011, confirming that national prevention progress has not been limited to select geographic areas. Prevention success was slightly less widespread in wards and NICUs, although progress was evident in the majority of jurisdictions for these locations as well. Furthermore, most of these jurisdictions reported accelerated prevention success in 2011 compared to 2010.

A major consideration for interpretation of these data and for future reports is assessing the confidence in the validity of the data reported. Completion of validation studies of CLABSI data was reported from 25 states during 2011 (up from 21 in 2010); evaluations included data quality assessment of missing or implausible values and/or detection of outlier facilities (e.g., number of infections, rates, denominators) in all 25 states, and audit of medical records in 14. Such validation studies occurred for CAUTI in 8 states, and for SSI in 15 states. All states provided information about any HAI validation activities that they have performed. Some states without mandatory reporting of a given HAI have performed validation on NHSN data that are voluntarily shared with them by facilities. Validation efforts by state departments of health represent an important step toward a more complete understanding of the HAI data reported to NHSN.

Regardless of the success of validation efforts, inherent variability in case-finding of HAIs will occur between facilities, explaining some of the differences in observed infection rates and facility-specific SIRs. Several efforts are in place to improve the accuracy and confidence in these HAI data. Web-based NHSN surveillance training modules are now available (http://www.cdc.gov/nhsn/training.html), which include webinars, slidesets, and self-paced, interactive, online training courses with continuing education credits available upon successful completion of an assessment. NHSN training is regularly provided during CDC-hosted events and at professional meetings and conferences. Improvements to the NHSN system to improve data accuracy continue to be made, including business rules and cross-field edit checks to prevent data entry errors, system alerts to inform users of missing data, and data quality reports to inform users of aberrant data.

As part of the National Action Plan to Prevent Healthcare-Associated Infections that was established in 2008, HHS has set goals for reducing CLABSI, CAUTI, and SSI by December 2013 (16). The data included in this report indicate that steady progress is occurring towards the goal of a 50% reduction in CLABSI over the course of 5 years (we report a 41% reduction from baseline in the third year) and the 25% reduction goal for SSI (we report a 17% reduction from baseline in the third year). Progress towards the 25% reduction goal for CAUTI is moving more slowly, with a 7% reduction from baseline in 2011 (this is the second year of measurement with a baseline year of 2009), but with sustained prevention efforts, the 2013 goal remains attainable.

The SIRs summarize complex data related to HAIs in a single set of indicators that use national data for a specified time period as a common referent group. The indirect standardization technique used to calculate SIRs is also used in the calculation of standardized mortality ratios (SMRs), a commonly used method in epidemiology for comparing mortality between a group and a referent population. This summary measure should not be used to derive any absolute ranking of facilities, states, or regions, but rather as a tool to identify facilities, states, or regions that may deserve targeted evaluations, which may include validation efforts or assessing potential prevention programs.

As more data is now available, improved methods of risk adjusting HAI data are being explored, including direct standardization of data reported comprehensively, the use of reliability adjusted SIRs, and additional measures of CAUTI prevention (such as a patient-day based rate). Measuring progress and performance from a single surveillance system has inherent challenges that we are committed to overcoming. Future reports will incorporate these new developments as we continue to explore the value and feasibly of applying new methods and operations to NHSN surveillance methodology and analysis.

Top of page

Conclusion

This report presents a set of national summary statistics for CLABSIs, CAUTIs, and SSIs for 2011, including serial SIRs for CLABSI, CAUTI, and SSI for 2010-2011. As a single summary measure of prevention success, there has been a large reduction (41%) in CLABSIs among reporting hospitals compared to predictions, with more modest reductions seen for CAUTI (7%) and SSI (17%). Prevention success improved in 2011 compared to 2010 for CLABSI. For SSI, improved prevention success over the two years was documented among five of the nine operative procedures evaluated, but the impact of new reporters in 2011 greatly influenced this observation. Overall, there is still substantial opportunity for improvement across a range of operative procedures. Additional progress can be made in CAUTI prevention, for which most of the national prevention success was limited to ward locations. Analyses using the CLABSI SIR at the state level, including serial comparisons of SIRs, provide a method for monitoring the impact of interventions and assessing the success of state-based and national HAI reduction efforts. As SSI and CAUTI reporting becomes more comprehensive in 2012, future SIR reports will include state-specific metrics for these HAIs as well. Ongoing interactions with state health departments will be critical in determining ways to improve the reporting of HAIs and ways to act on these data to prevent HAIs. The remaining burden from these HAIs, in terms of both numbers of infections (and the implicit associated morbidity and mortality) and increased reimbursements attributable to these infections highlights the ongoing need for HAI prevention as well as the data required to support such prevention. Publication of this report is one step among many in providing data needed for analysis and action at all levels, with the intent of spurring additional progress toward HAI elimination throughout the United States.

Top of page