Risk factors for Clostridium difficile cytotoxin-positive diarrhea after control for horizontal transmission.

OBJECTIVE

This study identifies risk factors that predispose hospitalized patients to diarrhea caused by Clostridium difficile.

DESIGN

Unlike most previous studies, this case-control study accounts for horizontal (person-to-person) transmission of this infectious organism by matching on location.

SETTING

An urban community hospital of 600 beds with both primary care and referred patients.

PATIENTS

Both the 36 cases and the 36 controls came from billing printouts for C difficile cytotoxin assays. Exclusion criteria included out-patient specimens and stays of 38 days or more. Controls had negative assays and were individually matched on the basis of the floor and by the general date of specimen collection.

RESULTS

Cases had a longer mean duration of antibiotic use (6.7 versus 4.1 days, p = .006). Of the other 22 factors that were evaluated for their risk of predisposing to C difficile, only clindamycin (OR = 3.50, p = .09) and third-generation cephalosporins (OR = 3.00, p = .04) showed any association. The odds ratio for third-generation cephalosporins in the absence of clindamycin was 3.50 (p = .09). Data were collected by physicians who were not blinded.

CONCLUSIONS

This study found that the clindamycin and third-generation cephalosporins were risk factors for C difficile-associated diarrhea after controlling for horizontal transmission. First-generation cephalosporins were not risk factors (OR = 0.86), while sample size considerations prevented firm conclusions about second-generation cephalosporins (OR = 1.67, p = .23). Many of the previously alleged risk factors might be explained by the horizontal transmission of an infectious disease; for instance, this seems a better explanation for an outbreak on a surgery ward than does surgery itself.