Accuracy of Catheter Associated Bloodstream Infection (CLABSI) Rates Based on Extrapolation of Monthly Central Line Days Collected One Day a Week

Background: Collecting central line days is a challenge at most healthcare facilities. In the absence of automation, infection preventionists or other staff manually count the number of patients with central lines daily with missed days often estimated. This process is impractical and resource consuming as the scope of infection prevention programs continue to expand.

Objective: To evaluate the feasibility of using a sample method of one week day to obtain central line days and how it impacts central line associated bloodstream infection (CLABSI) rates.

Methods: The number of patients with one or more central lines was collected on week days between July and September 2010 in 6 patient care units at 6 hospitals. The daily counts were totaled each month for each unit to obtain the “true central line days”. “Calculated central line days” were obtained by using the number of central line days on one day per week (excluding Saturdays and Sundays) and extrapolating this number for the full month. Percent errors were calculated to compare the two methods. The CLABSI rates for the NHSN 75^th percentile by unit type were computed by dividing the expected number of infections by the calculated central line days. Differences and percent errors were computed comparing the CLABSI rates using calculated central line days and the NHSN rate. Averages of all percent errors were computed to measure the difference between the two methods.

Results: For the study period, the average mean absolute percent error and average median absolute percent error comparing the true and calculated central line days was 10% and 6% respectively. Furthermore, the differences in denominators did not have a significant impact on the mean and median absolute percent error of the CLABSI rates (10% and 6% respectively).

Conclusions: Acquiring central line days to compute infection rates is a challenge but is necessary for healthcare associated infection (HAI) data submission to the NHSN. Collecting central line day data once per week and utilizing a calculation to acquire the total monthly central line days is a viable alternative to collecting data daily. In this study, we demonstrate that there is small variability in the mean and median of monthly central line days and CLABSI rates when the two methods are compared. Consequently, using this method may reduce the valuable resource burden of HAI data collection and further shift these resources to prevention activities.