National and State Healthcare-associated Infections Standardized Infection Ratio Report

Using Data Reported to the National Healthcare Safety Network

January – December 2010

Results

Table 1 (Tables 1a, 1b, 1c) summarize the variability and extent of HAI reporting to NHSN for each CLABSI, CAUTI, and SSI by state, respectively. CLABSI data were reported from at least one facility in 49 states and Washington, D.C., and in only seven of these 49 did fewer than five facilities contribute CLABSI data. In many instances a large number of facilities reported data in states without mandates, and overall 2,403 facilities contributed CLABSI data in 2010 compared to 1,695 (50% increase) in 2009. These facilities reported CLABSI data from 8,904 different locations (3,760, 42% critical care; 4,215, 52% ward; 529, 6% NICU). CAUTI data were reported from at least one facility in 47 states and Washington, D.C., and in only 13 of these 47 did fewer than five facilities contribute data. Overall 1,097 facilities contributed CAUTI data in 2010. These facilities reported from 4,193 different locations (1,491, 36% critical care; 2,702, 64% ward). SSI data were reported in 45 states and Washington, D.C., from 1,385 facilities (an increase of 46% from the 946 facilities reporting in 2009). Ten states had fewer than five facilities reporting SSI data. Overall, 529,038 surgical procedures were reported in 2010 compared to 420,340 during 2009 (25% increase).

Table 2 displays metrics summarizing the HAI experience for the United States. The first overall CLABSI measures include all patient care locations including non-neonatal patient care locations (critical care and wards as defined in the methods) and NICUs; during 2010, 13,812 CLABSIs were reported compared to 20,184.815 predicted for an SIR of 0.684 (95% CI 0.673-0.696). This translates to about a 32% national reduction compared to the referent period. Individual facilities reported a wide range of facility-specific SIRs; half of all facilities reported an SIR <.56 (the median), and 90% of facilities reported SIRs <1.52. This represents an improvement compared to the previous report where about 10% of facilities were reporting SIRs > 1.99. When stratified by patient care area groupings, the SIRs were lowest among non-neonatal critical care locations (SIR 0.654), then NICUs (SIR 0.695), followed by wards (SIR 0.728). The SIR for NICUs is improved compared to the previous report of July-December 2009, where the SIR was reported as 0.86 (95% CI 0.80- 0.93).4 Of note, 14 facilities reported only data from the newly defined mixed acuity locations for which there are no comparisons in the referent time period to calculate SIRs; these facilities were excluded from analysis in Table 2, resulting in 2,389 facilities contributing CLABSI data to the overall CLABSI SIR.

For CAUTIs reported from all patient care areas (excluding NICUs), 9,995 CAUTIs were reported while 10,656.872 were predicted, resulting in an SIR of 0.938 (95% CI 0.920-0.993). A slightly higher SIR was observed among critical care locations (SIR 0.967, 95% CI 0.942-0.993) compared to ward locations (SIR 0.903, 95% CI 0.876-0.930). This translates into a reduction in CAUTIs of about 3% (ICUs) to 10% (ward locations) since 2009 (the referent period for CAUTI).

The national SSI SIR was summarized across the procedure types outlined previously and was limited to SSIs classified as deep incisional or organ/space infection and detected during admission or readmission to the same hospital in which the procedure was performed. For the overall national SSI SIR, 4,737 deep incisional or organ/space SSIs were reported during initial admission or upon readmission from 529,038 procedures; based on the various factors reported for these procedures, 5,170.309 would have been predicted (SIR 0.916, 95% CI 0.89-0.943). Nationally, this experience translates to an 8% reduction in the incidence of these SSIs among this group of procedures, similar to what was reported in the July-December 2009 SIR report.⁴ Again, the facility-specific SIRs summarized in Table 2 demonstrate great variability; 25% of the facilities reported an SIR >1.29 (75th percentile), that is, 29% more SSIs than would have been predicted, similar to the experience reported last year.

When procedure-specific SIRs were calculated, the number of procedures reported within each category was a small subset of the total. Also, the number of facilities contributing data to any of the procedure-specific SIRs varied considerably. This included lows of 21, 31, and 50 facilities reporting data on rectal surgery, abdominal aortic aneurysm repair, or peripheral vascular bypass surgery respectively, to a high of 966 reporting data on knee arthroplasty. The resulting procedure-specific SIRs range from 0.648 to 1.285. However, only three of the procedure-specific SIRs significantly differed from 1.0: knee arthroplasty (SIR 0.892, 95% CI 0.840-0.947), coronary artery bypass graft surgery (SIR 0.820, 95% CI 0.766-0.876), and colon surgery (SIR 0.909, 95% CI 0.853-0.968). This may be in part due to small sample sizes of some procedure-specific estimates or lack of successful prevention efforts for any of a variety of reasons.

Table 3 illustrates state-specific SIR. For the overall CLABSI SIR calculations (Table 3a) in 42 states and Washington, D.C., SIR could be calculated. In 35 of these sites the SIR was significantly lower than 1.0. In roughly half of these sites, over 20 facilities reported enough data to calculate a reliable facility-specific SIR, allowing some assessment of the variability in performance across a wide range of facilities. In several states, >10% of individual facilities have considerably high SIRs (e.g., >1.9). Critical care location-specific (Table 3b), ward-specific (Table 3c), and NICU-specific (Table 3d) SIRs illustrate a similar pattern, although these estimates are less precise due to fewer data in each strata. Of note, validation activities including an external authority performing a medical record audit (YESa in Table 1) occurred in 16 states reported in Table 3b, mostly among states with a mandate to report to NHSN. The SIRs from states reporting such validation reported summary SIRs (Table 3b) modestly higher than other states; however none were >1.0, one was no different than 1.0, and 15 were significantly <1.0. The median SIR among states reporting such validation efforts was 0.675 compared to the value of 0.654 for the entire United States.

Table 4 presents serial SIRs for specific states with sufficient data in both reporting periods, comparing 2009 to 2010. SIRs represent CLABSI from all locations. The columns under "All Reporters" include data from all facilities reporting in either of the reporting periods, while those under "Continuously Reporting Locations" represent data from only those locations and facilities reporting in both years. Of the 52 reporting sites, nine had insufficient data to report serial SIRs. Of the remaining 43, only two reported increased SIRs, neither of which was statistically significant when restricted to continuously reporting facilities. In contrast, 21 reported decreases in CLABSI SIR, of which 20 remained significant when restricted to continuously reporting locations.

Table 5 presents serial SIRs for national CLABSI and SSI data for 2009 compared to 2010. These data assess progress in preventing HAIs between two sequential reporting periods. SIRs for "All Reporters" include data reported from non-NICU critical care locations, wards, and NICUs. For CLABSI, the SIR significantly decreased between reporting periods, indicating improved reductions compared to 2009. This finding was confirmed when evaluating only those locations reporting in both periods. For SSI, serial SIRs were significantly lower in 2010 compared to 2009 for the overall SSI SIR of coronary artery bypass graft surgery and rectal surgery. However, this measure of improved prevention success was confirmed for only the latter two scenarios. Regarding rectal surgery, the metric is based on only the 11 facilities reporting during the two-year period. The remaining procedures, and all procedures combined, had similar SIRs between the two years, but only knee arthroplasty, colon surgery, and coronary artery bypass graft surgery have SIRs <1.0 in 2010 and 2009 translating to successful reduction in SSIs (Table 4) compared to the baseline period.