![[ Visit SHEA online ]](images/banner.gif)
Background: Enzyme linked immunoassay (EIA) testing is insensitive for the diagnosis of C. difficile (CD), and recent studies suggest that its accuracy may be strain sensitive. In our hospital, CD-EIA was <50% sensitive when compared against the gold standard of CD stool culture. Thus, it became common practice for physicians confronted with patients with continued diarrhea and negative EIA reports to question these results. Typically, these patients were maintained on CD isolation and repeat testing was sent while the patient was initiated and continued on empiric treatment. In July 2010, our institution ceased EIA testing and began testing for the presence of CD via polymerase chain reaction (PCR).
Objective: To track the change in monthly number of cases and rates of CD, as well as the change in total CD-related isolation days before and after the introduction of CD-PCR testing.
Methods: Single center retrospective pre-post observational study.
Results: Our average CD rate pre-PCR was 7.8/10,000 patient-days vs. 15.4/10,000 patient-days post-PCR, representing an increase of 49%. Our average monthly number of CD cases was 7.7 pre-PCR, which rose to 14.3 post-PCR; an 86% increase. In contrast, our average monthly CD-related isolation days decreased from 340 to 181 between the pre-and the post-PCR periods, representing a decrease of 47% (Figure 1).
Conclusions: After the institution of PCR, despite a nearly doubling of the monthly CD rate and monthly CD case load, the total monthly number of CD isolation days decreased by almost half. We believe this paradoxical finding reflects two major changes: 1) the post PCR monthly increase in CD cases and rates reflects a detection bias introduced by more sensitive testing methods and not a true increase in CD disease, and 2) there was a marked change in physician behavior when provided accurate data (as physicians trust a negative PCR result they discontinue isolation more promptly). We hypothesize that the decreased number of CD isolation days might be associated with downstream financial benefits, such as decreased use of personal protective equipment, decreased use of unnecessary empiric therapy, and decreased length of stay. In addition to improving quality of care, we expect that these benefits would far exceed the increased costs of testing. Additional work is needed to confirm these preliminary findings and to establish their economic implications.
