Computer-Assisted Surveillance: Does It Work?

ISSUE: Traditional surveillance for nosocomial infections is labor-intensive and time-consuming. In today's healthcare setting, infection control professionals (ICPs) must be continuously searching for time-efficient alternative ways to collect reliable data. We sought to determine whether electronic review of our clinical-financial database could identify post–cesarean section (C-section) endometritis with sufficient accuracy to provide sound data to the obstetrics (OB) service. PROJECT: Our Labor and Delivery Department was asked to provide a monthly list of C-section patients, ASA scores, wound classifications, and durations of surgery. A Performance Improvement Specialist used DRGs to identify C-section patients and then ICD9 codes to identify cases of endometritis. Validity of the electronic data was confirmed by chart for the first 2 months. RESULTS: During the early phases, we found that chorioamnionitis was also captured. This was eliminated by changing the ICD9 screening codes. Other infections being captured under this code which were discounted were urinary tract infections and were due to coding errors. Electronic versus traditional surveillance initially showed only a 79% match. After changing our method of collecting the ICD9 codes, the match was 88%. The OB service now receives data they can benchmark against National Nosocomial Infection Surveillance (NNIS) rates. Providing them with this information has allowed them to evaluate their service and identify tools to improve the quality of patient care in our facility. LESSONS LEARNED: With a little effort up front, surveillance time can be greatly reduced without compromising the quality of the data received. Computer-assisted surveillance can be a valuable tool for today's ICP, but the success of this method is dependent on the accuracy of chart documentation and the experience of the coders.