Catheter-associated urinary tract infections (CAUTIs) account for >30% of hospital-acquired infections (HAIs) reported by acute-care hospitals.Reference Magill, Edwards and Bamberg 1 , Reference Lo, Nicolle and Coffin 2 Acute-care hospitals are incentivized to reduce CAUTIs because it is one of the measures included in the Centers for Medicare and Medicaid (CMS) hospital-acquired condition reduction program. 3 The National Healthcare Safety Network (NHSN) provides standardized criteria for the surveillance definitions for CAUTI. There were major concerns with the previous 2013 NHSN CAUTI surveillance definition.Reference Dicks, Baker and Durkin 4 Effective January 2015, significant changes were made to the NHSN CAUTI definition: (1) the removal of urinalysis criteria, (2) an increase in the urine culture bacterial threshold from 103 to 105 colony-forming units (cfu), and (3) the exclusion of yeasts or molds as potential CAUTI pathogens. 5 The objective of our study was to determine the impact of the current 2015 NHSN CAUTI definition on publicly reported CAUTI rates in intensive care units (ICUs) at our academic medical center.
METHODS
We performed a retrospective analysis of the prospectively collected CAUTI surveillance data from January 1, 2013, to June 30, 2016. The setting included 7 ICUs at the University of Alabama at Birmingham Hospital, a 1,157-bed academic medical center. Trained infection preventionists perform CAUTI surveillance using the applicable NHSN definition and calculate the standardized infection ratio (SIR). To decrease the incidence of CAUTI, a CAUTI prevention bundle was implemented in late 2013, which included a nurse-driven urinary catheter removal protocol, an annual mandatory HAI prevention education module for all healthcare providers, and training of nursing staff in the proper techniques for urinary catheter insertion.
We examined the trend of our reported CAUTI rates from January 2013 to June 2016 in 7 ICUs and applied the current 2015 CAUTI definition to 2013 and 2014 CAUTI cases. Rates were compared using Pearson’s χ2 test; means were compared using 2-sample t test; and P≤.05 was considered statistically significant. Catheter utilization ratio (CUR, catheter days divided by patient days) was calculated to determine changes in the volume of catheter use. Data analyses were performed using Stata version 12.0 (StataCorp, College Station, TX).
RESULTS
When the corresponding NHSN definition for the respective year was applied, we observed a trend for decreasing yearly CAUTI rates. Even before the NHSN definition was updated, but during the implementation of the CAUTI prevention bundle, we observed a significant decrease in the CAUTI incidence rate (IR) from 5.7 UTIs per 1,000 catheter days in 2013 to 3.9 UTIs per 1,000 catheter days in 2014 (P<.001). During the 2-year period between January 2013 and December 2014, 345 CAUTIs occurred, but more than half of these did not meet the current (2015) NHSN definition. Notably, 44.1% of CAUTIs in 2013 and 50.3% in 2014 were due to yeast (Table 1).
TABLE 1 Catheter-Associated Urinary Tract Infection (CAUTI) Rates by Previous (2013) and Current (2015) National Healthcare Safety Network (NHSN) Definitions
NOTE. IR, incidence rates (ie, IRs reported per 1,000 catheter days); UA, urinalysis; CUR, catheter utilization ratio (ie, catheter days/patient days).
a UA criteria: positive urinalysis plus bacterial threshold between 103 and 105 colony-forming units.
b Up to June 30, 2016.
With the current 2015 NHSN definition, we observed a significant decline in the CAUTI IR in 2015: 0.98 UTIs per 1,000 catheter days in 2015 versus 3.89 in 2014 (P<.001). When the current 2015 NHSN definition was applied to the 2013 CAUTI data, the CAUTI IR decreased by 57.5% from 5.7 UTIs per 1,000 device days using the old definition to 2.4 UTIs per 1,000 device days with the current definition (P=.001). Similarly, the yearly CAUTI IR decreased from 3.89 UTIs per 1,000 device days with the old definition to 1.5 UTIs per 1,000 device days with the current definition (P=.001, relative risk reduction of 61.2%). Even using the current 2015 definition, we observed a trend for decreasing CAUTI rates from 2013 to 2016 unrelated to the definition change: (2.4 UTIs per 1,000 catheter days in 2013 to 0.87 in 2016 (P<.001). We concurrently noted a decreasing trend in CUR from 0.82 in 2013 to 0.72 in 2016 (P=.09), but more urine cultures were done in 2016 (50 cultures per month) than in 2013 (36 cultures per month).
DISCUSSION
We found that the current 2015 NHSN CAUTI definition resulted in a >50% decline in reportable CAUTIs, but this result coincided with the positive effect of our CAUTI prevention bundle. Our experience underscores the importance of using epidemiologic definitions that closely correlate with clinical definitions for CAUTI, especially in the era of public reporting. Clinicians may become less engaged in infection prevention efforts if they perceive that the cumulative CAUTI SIR is not at goal because of a faulty definition. Prior to 2015, the NHSN CAUTI definition was too broad to accurately detect the true incidence of CAUTI. When the previous CAUTI definition was introduced in 2013, certain changes did not allow for exclusion of fever from the criteria even if fever could be attributed to an alternative diagnosis. Neelakantha et alReference Neelakanta, Sharma and Kesani 6 reported that this minor change in the definition led to a >2-fold increase in NHSN-defined CAUTIs in their study. The current 2015 NHSN definition is more specific and correlates more closely with the clinical definition. Our findings also suggest that the overall decline in CAUTI rates was, in part, due to the CAUTI prevention bundle with a corresponding decrease in CUR. Despite implementing these interventions, the cumulative CAUTI SIR in the ICUs was still >1 until the NHSN definition was changed.
In conclusion, accurate surveillance definitions that coincide with clinical definitions are essential because CAUTI rates are now publicly reported and may affect a hospital’s reimbursement rate. The definition change represents an opportunity to educate clinicians regarding these epidemiologically relevant definitions and appropriate use of antimicrobials.Reference Press and Metlay 7 Lastly, although the current cumulative CAUTI SIR is low, future interventions should target CUR and urine culture stewardship.Reference Mullin, Kovacs and Fatica 8
ACKNOWLEDGMENTS
We thank the members of the UAB Infection Prevention team for their support of this study.
Financial support: No financial support was provided relevant to this article.
Potential conflicts of interest: All authors report no conflicts of interest relevant to this article.
