Challenges and proposed improvements for reviewing symptoms and catheter use to identify National Healthcare Safety Network catheter-associated urinary tract infections

Background Retrospective medical record review is used to categorize urinary tract infections (UTIs) as symptomatic, catheter-associated, and/or healthcare-associated to generate National Healthcare Safety Network (NHSN) surveillance and claims data. We assessed how often patients with UTI diagnoses in claims data had a catheter in place, had documented symptoms, or met the NHSN criteria for catheter-associated UTI (CAUTI). Methods Two physicians retrospectively reviewed medical records for 294 randomly selected patients hospitalized with UTI as a secondary diagnosis, discharged between October 2008 and September 2009 from the University of Michigan. We applied a modification of recent NHSN criteria to estimate how often UTIs in claims data may be an NHSN CAUTI. Results The 294 patients included 193 women (66%). The mean patient age was 63 years, and the median length of hospital stay was 7.5 days. Catheter use was noted for 216 of 294 postadmission records (74%), including 126 (43%) with a Foley catheter. NHSN symptoms were noted in 113 records (38%); 62 (21%) had symptoms other than fever. Of 136 hospitalizations meeting urine culture criteria, 17 (5.8%) met the criteria for a potential NHSN CAUTI. Conclusions Retrospective medical record review to identify symptoms and catheter use is complicated and resource-intensive. Requiring standard documentation of symptoms and catheter status when ordering urine cultures could simplify and improve CAUTI surveillance and its fidelity as a hospital quality indicator.