Improved detection of orthopaedic surgical site infections occurring in outpatients

To test the hypothesis that a simple method of text searching of standard dictated outpatient clinic notes could be used to detect postoperative infections that otherwise were undetected by the traditional hospital-based infection surveillance methods, a retrospective analysis of all clinical notes from the orthopaedic surgery service of a tertiary care medical center without an electronic medical record system was done. Text-searching algorithms were used to screen all outpatient clinic notes for patients seen between September, 2000 and May, 2001. Clinic notes were screened using specific search terms indicative of surgical site infection, to yield a subset of notes that then were manually reviewed. The identified 18,887 notes were reduced to 558 screen-positive notes, of which 347 notes (197 patients) had confirmed surgical site infections on manual review (positive predictive value, 62.2%). Eighty percent of the total joint infections and 54% of the spine infections were not detected through standard hospital-based surveillance. This simple, inexpensive, method for screening clinic notes is effective in improving detection of postdischarge surgical site infections (and, therefore, quality assessment for surgery). It can be implemented in most current clinical settings.

Diagnostic study, Level II-1 (development of diagnostic criteria on basis of consecutive patients--with universally applied reference gold standard). See the Guidelines for Authors for a complete description of levels of evidence.