[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address Please contact the publisher to request reinstatement.
[Skip to Content Landing]
February 23, 2011

Surveillance Quality in Reporting Nosocomial Bloodstream Infection Rates

JAMA. 2011;305(8):779-780. doi:10.1001/jama.2011.157

To the Editor: Dr Lin and colleagues1 demonstrated the difficulties in standardizing the report of nosocomial bloodstream infection rates. The focus of the study was the inconsistencies in diagnosis of these infections, inter-institutional variability, and their consequences. How ever, another inherent bias in the current guidelines for the surveillance of central line–associated bloodstream infection (BSI) exists. This bias results from the methods of central line–associated BSI rate estimation, counting events in the numerator and central-line days in the denominator. Units with high counts of central-line days would tend to have lower rates as the denominator increases. The denominator is calculated by summing the number of patients in a unit who have at least 1 central venous catheter each day.2 Infected patients are not excluded. In practice, central venous catheters are not always removed once infection occurs, either because diagnosis of central line–associated BSI has not been established or because alternative venous access is not available. This bias is counterproductive, as one of the basic principles underlying preventive measures taken to decrease central line–associated BSI rates is to shorten the duration of central venous catheter placement.3