AN APPARENT EXCESS OF OPERATIVE SITE INFECTIONS - ANALYSES TO EVALUATE FALSE-POSITIVE DIAGNOSES

OBJECTIVE: To investigate an apparent excess of operative site infecti ons (OSI) reported according to doctor's diagnosis (presumptive OSI) b y applying objective criteria for classification (documented OSI). To examine potential consequences of habitual overdiagnosis of OSI. DESIG N: A case-control design was used to examine the clinical course of 18 case patients (12 presumptive OSI, six documented OSI) and 18 matched controls. Comparisons also were made between presumptive and document ed OSI patients. SETTING: A nonteaching community hospital. PATIENTS: Thirty-six patients having laminectomies done by the same surgeon. INT ERVENTION: Implementation of objective criteria for diagnosis of confi rmed OSI and reclassification of presumptive OSI patients. RESULTS: Po stoperatively, the frequency of specific adverse events within the ope rative site (including postoperative hematoma or bleeding; wound necro sis, dehiscence, or sinus tract; and dural tear) was 83% for documente d OSI patients, contrasted with 16.7% for presumptive OSI patients (P< .01) and controls (P=.007). Median days of inpatient stay were 27 for documented OSI, contrasted with 9.5 for presumptive OSI (P=.01) and 7 for controls (P<.001). CONCLUSION: Documented OSI patients were found to have significantly more adverse findings and longer lengths of stay than presumptive OSI patients or controls. The similarity of findings for presumptive OSI patients and controls suggests that the apparent excess frequency of OSI was caused by incorrect diagnosis. Whereas doc tor's diagnosis may be useful as an initial screen for OSI, use of obj ective criteria for confirming OSI may avert the consequences of overd iagnosis, including excessive length of stay and unnecessary therapy, which lead to elevated healthcare costs and threaten a physician's pra ctice (Infect Control Hosp Epidemiol 1995;16:712-716).